WO2006123918A1 - Query matching system and method, and computer readable recording medium recording program for implementing the method - Google Patents

Query matching system and method, and computer readable recording medium recording program for implementing the method Download PDF

Info

Publication number
WO2006123918A1
WO2006123918A1 PCT/KR2006/001890 KR2006001890W WO2006123918A1 WO 2006123918 A1 WO2006123918 A1 WO 2006123918A1 KR 2006001890 W KR2006001890 W KR 2006001890W WO 2006123918 A1 WO2006123918 A1 WO 2006123918A1
Authority
WO
WIPO (PCT)
Prior art keywords
query
search
search result
database
generating
Prior art date
Application number
PCT/KR2006/001890
Other languages
French (fr)
Inventor
Hyang Cheol Lee
Joon Hyock Moon
Bum Joo Park
Original Assignee
Nhn Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from KR1020050042470A external-priority patent/KR100623628B1/en
Priority claimed from KR1020050042481A external-priority patent/KR100709984B1/en
Application filed by Nhn Corporation filed Critical Nhn Corporation
Priority to JP2008512220A priority Critical patent/JP5027803B2/en
Publication of WO2006123918A1 publication Critical patent/WO2006123918A1/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/3331Query processing
    • G06F16/3332Query translation
    • G06F16/3338Query expansion
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/3331Query processing
    • G06F16/3332Query translation
    • G06F16/3335Syntactic pre-processing, e.g. stopword elimination, stemming

Definitions

  • the present invention relates to a method and system for matching a query with respect to a predetermined database including an index for a search, and more particularly, to a query match method and system capable of outputting a predetermined number of search results having high possibility of being matched with an original query by processing the original query via a plurality of processes according to various search logics.
  • a shopping mall relay service indicates a service of informing a goods purchaser about shopping malls that enter a predetermined relay website via a contract with a service provider. Accordingly, various types of information may be provided to the goods purchaser by classifying and introducing goods which are being sold in the shopping malls into each category or by comparing prices for each shopping mall with respect to identical goods.
  • the shopping mall relay service introduces goods of the shopping malls to the goods purchaser using various methods. When the goods purchaser selects particular goods which are being sold at a particular shopping mall, the shopping mall relay service connects the goods purchaser and the selected shopping mall, so that substantial sales of the goods may be completed in the affiliated shopping mall. Goods information which is registered to each shopping mall must be registered in the shopping mall relay server for the shopping mall relay service.
  • goods registration procedures must be performed, the procedures including: 1) registering goods information of goods that a goods registrant desires to sell in a shopping mall server, 2) transmitting the registered goods information from the shopping mall server to a shopping mall relay server, 3) a department section in charge of the shopping mall relay server, matching and arranging a goods model name of the transmitted goods information through a manual operation, and 4) registering the arranged goods information to the shopping mall relay server.
  • a ratio of unprocessed goods information may exceed the ratio of the registered goods information.
  • a database search system may only output a search result which had exactly matched an original query inputted from a user or another system.
  • the database search system cannot utilize a plurality of processes of various search logics on the original query to enable matching of the original query, to output a predetermined number of search results.
  • a conventional shopping mall relay server includes the described general database search system, (1) it is impossible to retrieve a matching goods model when a goods name of goods registered in an affiliated shopping mall server by a goods registrant is incorrect or a model name field of the goods is blank, (2) since a search is not performed using a keyword registered in a goods model, it is impossible to retrieve a matching goods model when a goods name is registered as a certain keyword, (3) since a search is not performed using a search engine, it is not easy to perform the search when using a Korean name for goods, and (4) it is impossible to retrieve a goods model when there is an incorrect or omitted character in the goods name of the goods registered in the affiliated shopping mall server by the goods registrant.
  • An aspect of the present invention provides a query matching method and system which can overcome a conventional database search method limitation of only providing a search result precisely matching an original query and can output a predetermined number of search results capable of being matched with the original query by processing the original query via a plurality of processes according to various search logics.
  • An aspect of the present invention also provides a query matching method and system in which an automated goods registration/match operation is embodied while registering goods information registered in a shopping mall in a shopping mall relay server, thereby notably reducing associated operation resource.
  • An aspect of the present invention also provides a query matching method and system in which different search logic is applied according to a property of goods corresponding to a respective query, thereby efficiently retrieving a goods model having high probability of being matched with a corresponding query.
  • An aspect of the present invention also provides a query matching method and system in which a suitable query is extracted from an original query that is a goods name inputted to a shopping mall server by a goods registrant, and a goods model matched with a corresponding goods is retrieved by using the extracted query, thereby pre-performing goods registration/match operation when the goods registrant registers a goods in a shopping mall server, and without additional manual operation for the goods registration/match when registering the goods in a shopping mall relay server.
  • a method of matching a query with respect to a predetermined database including an index for a search including: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; dividing the original query into a plurality of words according to space, searching the database by using a word formed of any one or a composite of at least two of alphabetic characters, numerals, special characters from the plurality of words, as a second query, and generating a second search result, when a number of the first search results is less than a predetermined reference value; searching the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, selected from the second queries, as a third query and generating a third search result, when a sum of the first and second search results is less than the predetermined reference value; and outputting at least one of the first search result, the second search
  • a method of matching a query with respect to a predetermined database including an index for a search including: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; dividing the original query into a plurality of words according to space, searching the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, from the plurality of words, as a second query, and generating a second search result, when a number of the first search results is less than a predetermined reference value; generating a plurality of tokens by dividing the second query into special characters, searching the database by using a third query generated by connecting the tokens by space, and generating a third search result, when a sum of the first and second search results is less than the predetermined reference value; and outputting at least one of the first search result, the second search result, and the third
  • a method of matching a query with respect to a predetermined database including an index for a search including: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; searching the database by using a second query generated by dividing the original query according to space and generating a second search result, when a number of the first search results is less than a predetermined reference value; and outputting the first search result or the second search result as a total search result matched with the original query.
  • FIG. 1 is a flowchart illustrating a query matching method according to an embodiment of the present invention
  • FIG. 2 is a flowchart illustrating a process of outputting a search result matched with an original query via various processing logics after dividing the original query into a plurality of tokens, according to an embodiment of the present invention
  • FIG. 3 is a diagram illustrating an example of a search field according to an embodiment of the present invention
  • FIG. 4 is a diagram illustrating an example of a search result view capable of being outputted by performing the query matching method according to the present invention
  • FIG. 5 is a diagram illustrating another example of a search result view capable of being outputted by performing the query matching method according to the present invention
  • FIG. 6 is a flowchart illustrating a query matching method according to another embodiment of the present invention.
  • FIG. 7 is a diagram illustrating still another example of a search result view capable of being outputted by performing the query matching method according to the present invention.
  • FIG. 8 is a block diagram illustrating a configuration of a query matching system according to an embodiment of the present invention.
  • FIG. 9 is a block diagram illustrating a configuration of a query matching system according to yet another embodiment of the present invention.
  • FIG. 1 is a flowchart illustrating a query matching method according to an embodiment of the present invention.
  • the query matching method according to the present embodiment may be performed with respect to a predetermined database including an index for a search. Also, the query matching method may be performed by a predetermined query matching system.
  • the query matching system searches the database by using a first query generated by removing a space or a special character from an original query and generates a first search result.
  • the query matching system may generate the first query by removing ' ', V, '-', and '+' from the original query and may generate the first search query including the first query by searching the database.
  • the query matching system determines whether a number of the first search results is less than a predetermined reference value.
  • the predetermined reference value may be determined to be any one of 2 to 10.
  • the query matching system may determined whether to perform a next operation by determining whether the number of the first search results is less than 5 in S 102.
  • the first search results are outputted as a total search result in SI lO. However, when the number of the first results is less than the predetermined reference value, S 103 is performed.
  • the query matching system divides the original query into a plurality of words according to a space, searches the database by using a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters from the plurality of words, as a second query, and generates a second search result, when a number of the first search results is less than a predetermined reference value.
  • the query matching system may divide the original query by ' ', generate words formed of only alphabetic characters, numerals, and V, '-', and '+' as the second query, and search the database by using the second query, thereby generating the second search result.
  • the described process is performed on the assumption that English words are more important. Words determined to be unnecessary for searching the database according to the assumption are removed.
  • the query matching system determines whether a sum of the number of the first search results and a number of the second search results is less than the predetermined reference value.
  • the query matching system may determine a next operation to progress to by determining whether the sum of the number of the first search results and the number of the second search results is less than 5.
  • the first search results and the second search results are outputted as a total search result.
  • S 105 will be performed.
  • the query matching system searches the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral as a third query and generates a third search result.
  • the query matching system arranges a plurality of the third queries in a descending order of length and searches the database according to the order, thereby generating the third search result.
  • FIG. 2 is a flowchart illustrating a process of outputting a search result matched with an original query via various processing logics after dividing the original query into a plurality of tokens, according to an embodiment of the present invention.
  • S201 through S208 shown in FIG. 2 may be included in S 105 of FIG. 1.
  • the query matching system divides the third query into tokens formed of a group of only alphabetic characters or only numerals.
  • the query matching system may divide the third query such as DGB-020BE into a plurality of the tokens such as 1 DGB 1 , '020', and 'BE' which is formed of a group of only alphabetic characters or only numerals.
  • the query matching system generates a fourth query by connecting the tokens by a space such as ' '.
  • the query matching system repeatedly searches the database while decreasing a number of the tokens included in the fourth query from a latter part of the fourth query.
  • the query matching system may repeatedly search the database while decreasing the number of the tokens included in the fourth query from the latter part in an order of 'DGB 020 BE', 'DGB 020', and 1 DGB'.
  • the query matching system determines whether the tokens left while performing S203 are corresponding to a token formed of a group of alphabetic characters.
  • the left tokens are determined to be not corresponding to the token formed of the group of alphabetic characters in S204, in S206, whether a search result exists is determined.
  • S205 will be performed.
  • the query matching system searches the database by removing a space between the left tokens.
  • the query matching system may divide the third query such as KTF-X5500 into the plurality of the tokens formed of the group of only alphabetic characters or only numerals, such as 1 KTF', 1 X', and '5500' in S201, and may generate the fourth query such as 1 KTF X 5500' by connecting the tokens by space in S202, may search by 1 KTF X 5500' and 'KTF X' while decreasing the number of the tokens from the latter part of the fourth query in S203, may determine whether the left tokens such as 'KTF X' are the tokens formed of the group of alphabetic characters in S204, and may search the database by using 'KTFX' generated by removing the space between the left tokens in S205.
  • the third query such as KTF-X5500 into the plurality of the tokens formed of the group of only alphabetic characters or only numerals, such as 1 KTF', 1 X', and '5500' in S201
  • the fourth query such as 1 K
  • the query matching system determines whether a search result exists in the searching process of S205 in S206, and suspends the searching process when the search result exists in S207. However, when the search result does not exist, the searching process is repeatedly performed from S203.
  • the query matching system searches the database by using 'KTFX' in S205
  • the database may be repeatedly searched by decreasing the number of the tokens from the latter part, such as 'KTF X' and 'KTF', in S203.
  • the query matching system suspends the repeated searching process in S207 and generates the existing search result as the third search result in S208.
  • the query matching method in which the index for a search, included in the database, includes a plurality of search fields including a goods model name field and a keyword field and S105 of FIG. 1, including S201 through S208 shown in FIG. 2, is performed limited to the goods model name and keyword field of the index for a search.
  • FIG. 3 is a diagram illustrating an example of a search field according to an embodiment of the present invention.
  • the index for a search included in the database, includes a plurality of search fields including a goods model name field 301, a keyword field 302, a category field, a brand field, a price field, and a production date field.
  • S 105 of FIG. 1, including S201 through S208 shown in FIG. 2 may be performed limited to the goods model name field 301 and the keyword field 302 of the index for a search.
  • a matched goods model may not be retrieved.
  • the query matching system since the query matching system according to the present invention searches the keyword field 302, when a goods name is registered as a certain keyword, a matched goods model may be searched and outputted.
  • the query matching system determines whether a sum of the number of the first search results, the number of the second search results, and a number of the third search results is less than the predetermined reference value.
  • the query matching system may determine whether to perform a next operation by determining whether the sum of the number of the first search results, the number of the second search results, and a number of the third search results is less than 5 in S 106.
  • the first through third search results are outputted as the total search result.
  • S 107 will be performed.
  • the query matching system divides the original query into a plurality of tokens by using a predetermined search engine indexer, searches the database by using the tokens, and generates a fourth search result.
  • the query matching system determines whether a sum of the number of the first search results, the number of the second search results, the number of the third search results, and a number of the fourth search result is less than the predetermined reference value. When the sum is determined to be not less than the predetermined reference value in S 108, in SI lO 5 the first through fourth search results are outputted as the total search result. When the sum is determined to be less than the predetermined reference value, S 109 will be performed.
  • the query matching system generates a fifth search result including at least one of the tokens from the database. Namely, the query matching system performs a search with respect to all words including at least one token when the sum of the search results is less than the predetermined reference value.
  • the query matching system outputs at least one of the first through fifth search results as the total search result matched with the original query.
  • FIG. 4 is a diagram illustrating an example of a search result view capable of being outputted by performing the query matching method according to the present invention.
  • a matched goods model cannot be retrieved.
  • 401 shown in FIG. 4 illustrates a search result according to a conventional technology.
  • matching process can only fail when the model name filed is blank.
  • a goods name 402 corresponding to the original query is processed according to various search logics shown in FIGS 1 and 2 to perform search, a goods model exactly matching the goods name 402 may be retrieved as shown in 403.
  • the goods name 402 that is the original query is "[six monthly installments without interest/instant discount by 5%] portable multimedia player (PMP) AV340R (40GB)".
  • PMP portable multimedia player
  • S 105 the original query is divided by space, a word formed of only alphabetic characters, numerals, and special characters is extracted, and a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, becomes the third query. Accordingly, the third query generated by processing the goods name 402 in S 105 is determined as 'AV340R'.
  • the query matching system divides 'AV340R' that is the third query into tokens formed of a group of only alphabetic characters or only numerals in S201 and generates the fourth query by connecting the tokens by space in S202. Accordingly, the fourth query is determined to be 'AV 340 R'. Also, the query matching system repeatedly searches the database by decreasing the number of the tokens from the latter part of the fourth query in S203. Accordingly, the query matching system may repeatedly search the database in an order of 1 AV 340 R', 1 AV 340', and 'AV, when a search result exist while searching in S206, may suspend repeated searching in S207, and may generate the search result as the third search result in S208.
  • AV340(40G) is outputted in 403. It may be known that the query matching system searches the database by using 1 AV 340', recognizes that the search result in which the goods model name is "AV340(40G)" exists, suspends repeated searching, and outputs the search result on a search result view.
  • the model name field is blank, since the search is performed by processing the goods name 402 corresponding to the original query according to various search logics shown in FIGS. 1 and 2, a goods model matched with the goods name 402 may be precisely searched and outputted as shown in 403.
  • FIG. 5 is a diagram illustrating another example of a search result view capable of being outputted by performing the query matching method according to the present invention.
  • a goods model matched with the goods name 501 may be precisely searched and outputted as shown in 502.
  • the query matching method according to the various search logics shown in FIGS. 1 and 2 may be performed by a predetermined goods information registration system having a query matching system according to the present invention.
  • the index for a search, included in the database may include a plurality of search fields associated with a goods model, and the goods model may be a goods model determined to be an electric appliances/computer series.
  • FIG. 6 is a flowchart illustrating a query matching method according to another embodiment of the present invention.
  • the query matching method according to the present embodiment may be performed with respect to a predetermined database including an index for a search. Also, the query matching method may be performed by a predetermined query matching system.
  • the query matching system searches the database by using a first query generated by removing spaces or special characters from an original query, and generates a first search result.
  • the query matching system may generate the first query by removing ' ', 7, '-', and '+' from the original query and may generate the first search result including the first query by searching the database.
  • the query matching system determines whether a number of the first search results is less than a predetermined reference value.
  • the predetermined reference value may be determined to be any one of 2 through 10. For example, when the predetermined reference value is 5, the query matching system may determine whether to perform a next operation by determining whether the number of the first search results is less than 5 in S602.
  • the first search result is outputted as a total search result.
  • S603 will be performed.
  • the query matching system divides the original query into a plurality of words according to a space, searches the database by using a word whose first and last characters are alphabetic characters or numerals and including at least one numeral as a second query, and generates a second search result.
  • the query matching system may perform search by determining '1000002538' as the second query in S603.
  • the query matching system may perform search by determining '34-2821_esp' as the second query.
  • the query matching system may arrange a plurality of the second queries in a descending order of length, may search the database according to the order, and may generate the second search result.
  • the query matching system determines whether a sum of the number of the first search results and a number of the second search results is less than the predetermined reference value. For example, when the reference value is 5, the query matching system may determine whether to perform a next operation by determining whether the sum is less than 5 in S604.
  • the first search results and the second search results are outputted as the total search result.
  • S605 will be performed.
  • the query matching system divides the second query into a plurality of tokens by special characters, searches the database by using a third query generated by connecting the tokens by space, and generates a third search results.
  • the query matching system may generate the tokens by dividing the second query by V, '-', '+', '_', and V.
  • the query matching system may generate the third search result via a process of decreasing a number of the tokens included in the third query from a latter part of the third query while repeatedly searching the database.
  • the query matching system may repeatedly search the database in an order of '34 2821 esp 1 , '34 2821', and '34'.
  • the query matching system may repeatedly search the database while decreasing the number of the tokens included in the third query from the latter part in an order of 1 DGB 020 BE', 1 DGB 020', and 1 DGB'. Also, in this case, when it is determined that a search result exists while performing S605, the query matching system may suspend the repeated search process and may generate the existing search result as the third search result.
  • the query matching system outputs at least one of the first through third search results as the total search result matched with the original query.
  • the index for a search included in the database, may include a plurality of search fields including a goods model name field and a keyword field, and S601 through S606 shown in FIG. 6 may be performed limited to the goods model name field and the keyword field of the index for a search.
  • S 106 through SI lO described with reference to FIG. 1 may be additionally performed in the query matching method according to the present embodiment.
  • the query matching method according to the present embodiment may be performed by a predetermined goods information registration system having the query matching system according to the present invention
  • the index for a search, included in the database may include a plurality of search fields associated with a goods model, and the goods model may be included in a brand-name item series.
  • a very suitable search result may be outputted when performing the search according to the query matching method according to the present embodiment.
  • the query matching method according to the present embodiment may be performed with respect to a predetermined database including an index for a search and may be performed by a predetermined query matching system.
  • the query matching system searches the database by using a first query generated by removing a space or a special character from an original query and generates a first search result.
  • S602 is identical with the description of the previous embodiment.
  • the query matching system searches the database by using a word whose first and last characters are alphabetic characters or numerals and including at least one numeral as a second query and generates a second search result.
  • S604 is identical with the description of the previous embodiment.
  • the query matching system searches the database by using a third query generated by morphologically analyzing the original query and generates a third search results.
  • the query matching system outputs at least one of the first through third search results as the total search result matched with the original query.
  • the query matching method according to the present embodiment may be performed by a predetermined goods information registration system having the query matching system according to the present invention
  • the index for a search, included in the database may include a plurality of search fields associated with a goods model
  • the goods model may be included in cosmetics, powdered milk, diapers, or other general goods series.
  • a very suitable search result may be outputted when performing the search according to the query matching method according to the present embodiment.
  • FIG. 7 is a diagram illustrating still another example of a search result view capable of being outputted by performing the query matching method according to the present invention. Since the conventional shopping mall relay server having a general database search system and search is not performed by a search engine, it is not easy to search by using a Korean goods name. Since names of cosmetics and powdered milk are generally in Korean, it is very difficult to output a search result according to a conventional technology.
  • Korean goods name is performed by processing a goods name 701 corresponding to an original query according to various search logics of the present embodiment, a goods model matched with the goods name 701 may be precisely searched and outputted as shown in 702.
  • the query matching method according to the present invention may be embodied as a program instruction capable of being executed via various computer units and may be recorded in a computer-readable recording medium.
  • the computer readable medium may include a program instruction, a data file, and a data structure, separately or cooperatively.
  • the program instructions and the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those skilled in the art of computer software arts.
  • Examples of the computer-readable media include magnetic media (e.g., hard disks, floppy disks, and magnetic tapes), optical media (e.g., CD-ROMs or DVD), magneto-optical media (e.g., optical disks), and hardware devices (e.g., ROMs, RAMs, or flash memories, etc.) that are specially configured to store and perform program instructions.
  • the media may also be transmission media such as optical or metallic lines, wave guides, etc. including a carrier wave transmitting signals specifying the program instructions, data structures, etc.
  • Examples of the program instructions include both machine code, such as produced by a compiler, and files containing high- level language codes that may be executed by the computer using an interpreter.
  • the hardware elements above may be configured to act as one or more software modules for implementing the operations of this invention.
  • FIG. 8 is a block diagram illustrating a configuration of a query matching system 800 according to an embodiment of the present invention.
  • the query matching system 800 may perform a search process with respect to a predetermined database including an index for a search.
  • the query matching system 800 includes a first search unit 801, a second search unit 802, a third search unit 803, and an output unit 804.
  • the first search unit 801 searches the database by using a first query generated by removing spaces or special characters from an original query and generates a first search result.
  • the first search unit 801 may generate the first query by removing ' ', V, '-', and '+' from the original query and may generate the first search query including the first query by searching the database.
  • the second search unit 802 divides the original query into a plurality of words according to a space, searches the database by using a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters from the plurality of words, as a second query, and generates a second search result.
  • the second search unit 802 may divide the original query by ' ', generate words formed of only alphabetic characters, numerals, and '/', '-', and '+' as the second query, and search the database by using the second query, thereby generating the second search result.
  • the described process is performed on the assumption that English words are more important. Words determined to be unnecessary for searching the database according to the assumption are removed.
  • the third search unit 803 searches the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral as a third query and generates a third search result.
  • the third search unit 803 may arrange a plurality of the third queries in a descending order of length and may search the database according to the order, and may generate the third search result.
  • the third search unit 803 may perform S201 through S208 shown in FIG. 2.
  • the output unit 804 outputs at least one of the first search results, the second search results, and the third search results, as a total search result matched with the original query.
  • the query matching system 800 may operate being included in a predetermined goods information registration system, an index for a search, included in the database, may include a plurality of search fields associated with a goods model, and the goods model may be included in an electric appliances/computer series.
  • a very suitable search result may be performed when performing the search by the query matching system 800 according to the present embodiment.
  • the query matching system 800 may perform a search process with respect to a predetermined database including an index for a search and includes a first search unit 801, a second search unit 802, a third search unit 803, and an output unit 804.
  • the first search unit 801 searches the database by using a first query generated by removing spaces or special characters from an original query and generates a first search result.
  • the second search unit 802 divides the original query into a plurality of words according to a space, searches the database by using a word whose first and last characters are alphabetic characters or numerals and including at least one numeral as a second query, and generates a second search result. For example, when the original query is "Tag Heuher sunglasses
  • the second search unit 802 may perform search by determining '1000002538' as the second query.
  • the second search unit 802 may perform search by determining ' 34-282 l_esp' as the second query.
  • the third search unit 803 divides the second query into a plurality of tokens by special characters, searches the database by using a third query generated by connecting the tokens by a space, and generates a third search results.
  • the third search unit 803 may generate the tokens by dividing the second query by '/', '-', '+', '_', and '.'. Also, the third search unit 803 may generate the third search result via a process of decreasing a number of the tokens included in the third query from a latter part of the third query while repeatedly searching the database.
  • the third search unit 803 may repeatedly search the database in an order of '34 2821 esp 1 , '34 2821', and '34'.
  • the output unit 804 outputs at least one of the first search results, the second search results, and the third search results as a total search result matched with the original query.
  • the query matching system 800 may operate being included in a predetermined goods information registration system, an index for a search, included in the database, may include a plurality of search fields associated with a goods model, and the goods model may be included in a brand-name item series. Since a name of the goods model included in the brand-name item is generally formed of numerals or a combination of alphabetic characters and numerals, and Korean is generally used for only a description with respect to the goods, a very suitable search result may be performed when performing the search by the query matching system 800 according to the present embodiment.
  • the query matching system 800 may perform a search process with respect to a predetermined database including an index for a search and includes a first search unit 801, a second search unit 802, a third search unit 803, and an output unit 804.
  • the first search unit 801 searches the database by using a first query generated by removing spaces or special characters from an original query and generates a first search result.
  • the second search unit 802 searches the database by using a second query generated by dividing the original query into a plurality of words according to a space and generates a second search result.
  • the third search unit 803 searches the database by using a third query generated by morphologically analyzing the original query and generates a third search result.
  • the output unit 804 outputs at least one of the first search results, the second search results, and the third search results as a total search result matched with the original query.
  • the query matching system 800 may operate being included in a predetermined goods information registration system, an index for a search, included in the database, may include a plurality of search fields associated with a goods model, and the goods model may be included in cosmetics, powdered milk, diapers, or other general goods series. Since a name of the goods model included in the cosmetics, powdered milk, diapers, or other general goods series is generally in Korean, a more suitable search result may be performed when performing the search by the query matching system 800 according to the present embodiment, than the previous embodiments.
  • FIG. 9 is a block diagram illustrating a configuration of a query matching system 900 according to yet another embodiment of the present invention.
  • the query matching system 900 may perform a search process with respect to a predetermined database including an index for a search and includes a division unit 901, an extraction unit 902, a determination unit 903, a search unit 904, and an output unit 905 from elements shown in FIG. 9.
  • the division unit 901 divides an original query into a plurality of words according to a space.
  • the determination unit 902 determines a word formed of any one of or a composite of two or more of alphabetic characters, numerals, and special characters from the plurality of words, as a search query.
  • the search unit 904 searches the database by using the search query and generates a search result.
  • the output unit 905 outputs the search result as a search result matched with the original query.
  • the query matching system 900 may perform a search process with respect to a predetermined database including an index for a search and include a division unit 901, an extraction unit 902, a determination unit 903, a search unit 904, and an output unit 905 shown in FIG. 9.
  • the division unit 901 divides an original query into a plurality of words according to a space.
  • the extraction unit 902 extracts a word formed of any one of or a composite of two or more of alphabetic characters, numerals, and special characters, from the plurality of words.
  • the determination unit 903 determines a word whose first and last characters are alphabetic characters or numerals and including at least one numeral, from the extracted words, as a search query.
  • the search unit 904 searches the database by using the search query and generates a search result.
  • the output unit 905 outputs the search result as a search result matched with the original query.
  • a query matching system 900 according to still another embodiment of the present invention will be described with reference to FIG. 9.
  • the query matching system 900 may perform a search process with respect to a predetermined database including an index for a search and include a token generation unit 910 including a division unit 901 and an extraction unit 902, a determination unit 903, a search unit 904, and an output unit 905 shown in FIG. 9.
  • the token generation unit 910 generates a plurality of tokens by processing an original query according to predetermined division logic.
  • the token generation unit 910 may include the division unit 901 dividing the original query into a plurality of words according to a space and the extraction unit 902 extracting a word formed of any one of or a composite of two or more of alphabetic characters, numerals, and special characters, from the plurality of words, and may generate the plurality of tokens by dividing the extracted word into the plurality of tokens formed of a group of only alphabetic characters or only numerals or dividing the extracted word by special characters.
  • the determination unit 903 determines a search query by connecting the plurality of tokens by a space.
  • the search unit 904 repeatedly searches the database while decreasing a number of the tokens included in the search query from a latter part of the search query and generates a search result.
  • the output unit 905 outputs the search result as a search result matched with the original query.
  • a query matching method and system which can overcome a conventional database search method limitation of only providing a search result precisely matching an original query and can output a predetermined number of search results capable of being matched with the original query by processing the original query via a plurality of processes according to various search logics.
  • a query matching method and system in which an automated goods registration/match operation is embodied while registering goods information registered in a shopping mall in a shopping mall relay server, thereby notably reducing associated operation resource.
  • a query matching method and system in which different search logic is applied according to a property of goods corresponding to a respective query, thereby efficiently retrieving a goods model having high probability of being matched with a corresponding query.
  • a query matching method and system in which a suitable query is extracted from an original query that is a goods name inputted to a shopping mall server by a goods registrant and a goods model matched with a corresponding goods is retrieved by using the extracted query, thereby pre-performing goods registration/match operation when the goods registrant registers a goods in a shopping mall server, without additional manual operation for the goods registration/match when registering the goods in a shopping mall relay server.

Abstract

There is provided a query matching method and system for outputting a predetermined number of search results which can be matched with an original query by processing the original query via a plurality of processes according to various search logics. The query matching method with respect to a predetermined database including an index for a search includes: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; dividing the original query into a plurality of words according to space, searching the database by using a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters from the plurality of words, as a second query, and generating a second search result, when a number of the first search results is less than a predetermined reference value; searching the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, selected from the second queries, as a third query and generating a third search result, when a sum of the first and second search results is less than the predetermined reference value; and outputting at least one of the first search result, the second search result, and the third search result, as a total search result matched with the original query.

Description

QUERY MATCHING SYSTEM AND METHOD, AND COMPUTER
READABLE RECORDING MEDIUM RECORDING PROGRAM FOR
IMPLEMENTING THE METHOD
Technical Field
The present invention relates to a method and system for matching a query with respect to a predetermined database including an index for a search, and more particularly, to a query match method and system capable of outputting a predetermined number of search results having high possibility of being matched with an original query by processing the original query via a plurality of processes according to various search logics.
Background Art
Generally, a shopping mall relay service indicates a service of informing a goods purchaser about shopping malls that enter a predetermined relay website via a contract with a service provider. Accordingly, various types of information may be provided to the goods purchaser by classifying and introducing goods which are being sold in the shopping malls into each category or by comparing prices for each shopping mall with respect to identical goods. The shopping mall relay service introduces goods of the shopping malls to the goods purchaser using various methods. When the goods purchaser selects particular goods which are being sold at a particular shopping mall, the shopping mall relay service connects the goods purchaser and the selected shopping mall, so that substantial sales of the goods may be completed in the affiliated shopping mall. Goods information which is registered to each shopping mall must be registered in the shopping mall relay server for the shopping mall relay service. In a conventional art, goods registration procedures must be performed, the procedures including: 1) registering goods information of goods that a goods registrant desires to sell in a shopping mall server, 2) transmitting the registered goods information from the shopping mall server to a shopping mall relay server, 3) a department section in charge of the shopping mall relay server, matching and arranging a goods model name of the transmitted goods information through a manual operation, and 4) registering the arranged goods information to the shopping mall relay server.
Accordingly, a great amount of time is spent for the goods registration/matching according to the conventional art. Also, a limit of operation resources exists. As an example, a ratio of unprocessed goods information may exceed the ratio of the registered goods information.
Also, as a marketplace area is expanded, a continuous increase in goods is expected. Accordingly, when following the conventional goods registration procedure, the limit of operation resources may be further become more severe.
Also, generally, a database search system may only output a search result which had exactly matched an original query inputted from a user or another system. The database search system cannot utilize a plurality of processes of various search logics on the original query to enable matching of the original query, to output a predetermined number of search results.
However, since a conventional shopping mall relay server includes the described general database search system, (1) it is impossible to retrieve a matching goods model when a goods name of goods registered in an affiliated shopping mall server by a goods registrant is incorrect or a model name field of the goods is blank, (2) since a search is not performed using a keyword registered in a goods model, it is impossible to retrieve a matching goods model when a goods name is registered as a certain keyword, (3) since a search is not performed using a search engine, it is not easy to perform the search when using a Korean name for goods, and (4) it is impossible to retrieve a goods model when there is an incorrect or omitted character in the goods name of the goods registered in the affiliated shopping mall server by the goods registrant.
Disclosure of Invention Technical Goals
An aspect of the present invention provides a query matching method and system which can overcome a conventional database search method limitation of only providing a search result precisely matching an original query and can output a predetermined number of search results capable of being matched with the original query by processing the original query via a plurality of processes according to various search logics.
An aspect of the present invention also provides a query matching method and system in which an automated goods registration/match operation is embodied while registering goods information registered in a shopping mall in a shopping mall relay server, thereby notably reducing associated operation resource.
An aspect of the present invention also provides a query matching method and system in which different search logic is applied according to a property of goods corresponding to a respective query, thereby efficiently retrieving a goods model having high probability of being matched with a corresponding query. An aspect of the present invention also provides a query matching method and system in which a suitable query is extracted from an original query that is a goods name inputted to a shopping mall server by a goods registrant, and a goods model matched with a corresponding goods is retrieved by using the extracted query, thereby pre-performing goods registration/match operation when the goods registrant registers a goods in a shopping mall server, and without additional manual operation for the goods registration/match when registering the goods in a shopping mall relay server.
Technical Solutions
According to an aspect of the present invention, there is provided a method of matching a query with respect to a predetermined database including an index for a search, the method including: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; dividing the original query into a plurality of words according to space, searching the database by using a word formed of any one or a composite of at least two of alphabetic characters, numerals, special characters from the plurality of words, as a second query, and generating a second search result, when a number of the first search results is less than a predetermined reference value; searching the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, selected from the second queries, as a third query and generating a third search result, when a sum of the first and second search results is less than the predetermined reference value; and outputting at least one of the first search result, the second search result, and the third search result, as a total search result matched with the original query.
According to another aspect of the present invention, there is provided a method of matching a query with respect to a predetermined database including an index for a search, the method including: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; dividing the original query into a plurality of words according to space, searching the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, from the plurality of words, as a second query, and generating a second search result, when a number of the first search results is less than a predetermined reference value; generating a plurality of tokens by dividing the second query into special characters, searching the database by using a third query generated by connecting the tokens by space, and generating a third search result, when a sum of the first and second search results is less than the predetermined reference value; and outputting at least one of the first search result, the second search result, and the third search result, as a total search result matched with the original query.
According to still another aspect of the present invention, there is provided a method of matching a query with respect to a predetermined database including an index for a search, the method including: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; searching the database by using a second query generated by dividing the original query according to space and generating a second search result, when a number of the first search results is less than a predetermined reference value; and outputting the first search result or the second search result as a total search result matched with the original query.
Brief Description of Drawings
FIG. 1 is a flowchart illustrating a query matching method according to an embodiment of the present invention; FIG. 2 is a flowchart illustrating a process of outputting a search result matched with an original query via various processing logics after dividing the original query into a plurality of tokens, according to an embodiment of the present invention; FIG. 3 is a diagram illustrating an example of a search field according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating an example of a search result view capable of being outputted by performing the query matching method according to the present invention;
FIG. 5 is a diagram illustrating another example of a search result view capable of being outputted by performing the query matching method according to the present invention;
FIG. 6 is a flowchart illustrating a query matching method according to another embodiment of the present invention;
FIG. 7 is a diagram illustrating still another example of a search result view capable of being outputted by performing the query matching method according to the present invention;
FIG. 8 is a block diagram illustrating a configuration of a query matching system according to an embodiment of the present invention; and
FIG. 9 is a block diagram illustrating a configuration of a query matching system according to yet another embodiment of the present invention.
Best Mode for Carrying Out the Invention Hereinafter, a query matching method and system according to the present invention will be described in detail with reference to the attached drawings.
FIG. 1 is a flowchart illustrating a query matching method according to an embodiment of the present invention. The query matching method according to the present embodiment may be performed with respect to a predetermined database including an index for a search. Also, the query matching method may be performed by a predetermined query matching system.
In SlOl, the query matching system searches the database by using a first query generated by removing a space or a special character from an original query and generates a first search result. For example, the query matching system may generate the first query by removing ' ', V, '-', and '+' from the original query and may generate the first search query including the first query by searching the database. In S 102, the query matching system determines whether a number of the first search results is less than a predetermined reference value.
For example, the predetermined reference value may be determined to be any one of 2 to 10. When the predetermined reference value is 5, the query matching system may determined whether to perform a next operation by determining whether the number of the first search results is less than 5 in S 102.
When the number of the first search results is determined to be not less than the predetermined reference value in S 102, the first search results are outputted as a total search result in SI lO. However, when the number of the first results is less than the predetermined reference value, S 103 is performed.
In S 103, the query matching system divides the original query into a plurality of words according to a space, searches the database by using a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters from the plurality of words, as a second query, and generates a second search result, when a number of the first search results is less than a predetermined reference value.
For example, the query matching system may divide the original query by ' ', generate words formed of only alphabetic characters, numerals, and V, '-', and '+' as the second query, and search the database by using the second query, thereby generating the second search result. The described process is performed on the assumption that English words are more important. Words determined to be unnecessary for searching the database according to the assumption are removed.
In 104, the query matching system determines whether a sum of the number of the first search results and a number of the second search results is less than the predetermined reference value.
For example, when the predetermined reference value is 5, the query matching system may determine a next operation to progress to by determining whether the sum of the number of the first search results and the number of the second search results is less than 5. When the sum of the number of the first search results and the number of the second search results is determined to be not less than the predetermined reference value in S 104, in SI lO, the first search results and the second search results are outputted as a total search result. When the sum of the number of the first search results and the number of the second search results is determined to be less than the predetermined reference value in S 104, S 105 will be performed.
In S 105, the query matching system searches the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral as a third query and generates a third search result.
In this case, the query matching system arranges a plurality of the third queries in a descending order of length and searches the database according to the order, thereby generating the third search result.
FIG. 2 is a flowchart illustrating a process of outputting a search result matched with an original query via various processing logics after dividing the original query into a plurality of tokens, according to an embodiment of the present invention. S201 through S208 shown in FIG. 2 may be included in S 105 of FIG. 1. In S201, the query matching system divides the third query into tokens formed of a group of only alphabetic characters or only numerals.
For example, the query matching system may divide the third query such as DGB-020BE into a plurality of the tokens such as 1DGB1, '020', and 'BE' which is formed of a group of only alphabetic characters or only numerals. In S202, the query matching system generates a fourth query by connecting the tokens by a space such as ' '. In S203, the query matching system repeatedly searches the database while decreasing a number of the tokens included in the fourth query from a latter part of the fourth query.
For example, the query matching system may repeatedly search the database while decreasing the number of the tokens included in the fourth query from the latter part in an order of 'DGB 020 BE', 'DGB 020', and 1DGB'.
In S204, the query matching system determines whether the tokens left while performing S203 are corresponding to a token formed of a group of alphabetic characters. When the left tokens are determined to be not corresponding to the token formed of the group of alphabetic characters in S204, in S206, whether a search result exists is determined. When the left tokens are determined to be corresponding to the token formed of the group of alphabetic characters in S204, S205 will be performed.
In S205, the query matching system searches the database by removing a space between the left tokens.
For example, the query matching system may divide the third query such as KTF-X5500 into the plurality of the tokens formed of the group of only alphabetic characters or only numerals, such as 1KTF', 1X', and '5500' in S201, and may generate the fourth query such as 1KTF X 5500' by connecting the tokens by space in S202, may search by 1KTF X 5500' and 'KTF X' while decreasing the number of the tokens from the latter part of the fourth query in S203, may determine whether the left tokens such as 'KTF X' are the tokens formed of the group of alphabetic characters in S204, and may search the database by using 'KTFX' generated by removing the space between the left tokens in S205.
The query matching system determines whether a search result exists in the searching process of S205 in S206, and suspends the searching process when the search result exists in S207. However, when the search result does not exist, the searching process is repeatedly performed from S203.
In the previous example, though the query matching system searches the database by using 'KTFX' in S205, when the search result is determined not to exist in S206, the database may be repeatedly searched by decreasing the number of the tokens from the latter part, such as 'KTF X' and 'KTF', in S203.
While performing S203 through S205, when it is determined that the search result exist in S206, the query matching system suspends the repeated searching process in S207 and generates the existing search result as the third search result in S208.
According to an embodiment of the present invention, there is provided the query matching method in which the index for a search, included in the database, includes a plurality of search fields including a goods model name field and a keyword field and S105 of FIG. 1, including S201 through S208 shown in FIG. 2, is performed limited to the goods model name and keyword field of the index for a search.
FIG. 3 is a diagram illustrating an example of a search field according to an embodiment of the present invention.
Referring to FIG. 3, the index for a search, included in the database, includes a plurality of search fields including a goods model name field 301, a keyword field 302, a category field, a brand field, a price field, and a production date field.
Generally, when searching the database, all of the search fields of the index for a search have to be searched. However, according to an embodiment of the present invention, to improve search speed, S 105 of FIG. 1, including S201 through S208 shown in FIG. 2, may be performed limited to the goods model name field 301 and the keyword field 302 of the index for a search.
In a conventional shopping mall relay server having a general database search system, since a keyword registered in a goods model is not retrieved, when a goods name is registered as a certain keyword, a matched goods model may not be retrieved. However, since the query matching system according to the present invention searches the keyword field 302, when a goods name is registered as a certain keyword, a matched goods model may be searched and outputted.
As described above, the detailed processes of S 105 for generating the third search result has been described with reference to FIG. 2. Hereinafter, the present invention will be described with reference to FIG. 1.
In S 106, the query matching system determines whether a sum of the number of the first search results, the number of the second search results, and a number of the third search results is less than the predetermined reference value.
For example, when the predetermined reference value is 5, the query matching system may determine whether to perform a next operation by determining whether the sum of the number of the first search results, the number of the second search results, and a number of the third search results is less than 5 in S 106.
When the sum is determined to be not less than the predetermined reference value in S 106, in SI lO, the first through third search results are outputted as the total search result. When the sum is determined to be less than the predetermined reference value in S 106, S 107 will be performed.
In S 107, the query matching system divides the original query into a plurality of tokens by using a predetermined search engine indexer, searches the database by using the tokens, and generates a fourth search result. In S 108, the query matching system determines whether a sum of the number of the first search results, the number of the second search results, the number of the third search results, and a number of the fourth search result is less than the predetermined reference value. When the sum is determined to be not less than the predetermined reference value in S 108, in SI lO5 the first through fourth search results are outputted as the total search result. When the sum is determined to be less than the predetermined reference value, S 109 will be performed. In S 109, the query matching system generates a fifth search result including at least one of the tokens from the database. Namely, the query matching system performs a search with respect to all words including at least one token when the sum of the search results is less than the predetermined reference value.
In Sl 10, the query matching system outputs at least one of the first through fifth search results as the total search result matched with the original query.
FIG. 4 is a diagram illustrating an example of a search result view capable of being outputted by performing the query matching method according to the present invention.
In the conventional shopping mall relay server having a general database search system, when a name of goods, registered by a goods registrant via a shopping mall server, is not correct or a model name filed corresponding to the goods is blank, a matched goods model cannot be retrieved. 401 shown in FIG. 4 illustrates a search result according to a conventional technology. In the conventional technology as shown in 401, matching process can only fail when the model name filed is blank. However, according to the present invention, when the model name field is blank, a goods name 402 corresponding to the original query is processed according to various search logics shown in FIGS 1 and 2 to perform search, a goods model exactly matching the goods name 402 may be retrieved as shown in 403.
Namely, in FIG. 4, the goods name 402 that is the original query is "[six monthly installments without interest/instant discount by 5%] portable multimedia player (PMP) AV340R (40GB)". In S 105, the original query is divided by space, a word formed of only alphabetic characters, numerals, and special characters is extracted, and a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, becomes the third query. Accordingly, the third query generated by processing the goods name 402 in S 105 is determined as 'AV340R'.
Also, the query matching system according to the present invention divides 'AV340R' that is the third query into tokens formed of a group of only alphabetic characters or only numerals in S201 and generates the fourth query by connecting the tokens by space in S202. Accordingly, the fourth query is determined to be 'AV 340 R'. Also, the query matching system repeatedly searches the database by decreasing the number of the tokens from the latter part of the fourth query in S203. Accordingly, the query matching system may repeatedly search the database in an order of 1AV 340 R', 1AV 340', and 'AV, when a search result exist while searching in S206, may suspend repeated searching in S207, and may generate the search result as the third search result in S208.
Referring to FIG. 4, a search result in which a goods model name is
"AV340(40G)" is outputted in 403. It may be known that the query matching system searches the database by using 1AV 340', recognizes that the search result in which the goods model name is "AV340(40G)" exists, suspends repeated searching, and outputs the search result on a search result view.
As described above, according to the present invention, though the model name field is blank, since the search is performed by processing the goods name 402 corresponding to the original query according to various search logics shown in FIGS. 1 and 2, a goods model matched with the goods name 402 may be precisely searched and outputted as shown in 403.
FIG. 5 is a diagram illustrating another example of a search result view capable of being outputted by performing the query matching method according to the present invention.
In the conventional shopping mall relay server having a general database search system, when there is an incorrect or omitted word in a name of a goods, registered by a goods registrant via a shopping mall server, a matched goods model cannot be retrieved.
However, according to the present invention, when there is an incorrect or omitted word in a goods name, since a search is performed by processing a goods name
501 corresponding to the original query according to various search logics shown in FIGS. 1 and 2, a goods model matched with the goods name 501 may be precisely searched and outputted as shown in 502.
On the other hand, the query matching method according to the various search logics shown in FIGS. 1 and 2 may be performed by a predetermined goods information registration system having a query matching system according to the present invention. The index for a search, included in the database, may include a plurality of search fields associated with a goods model, and the goods model may be a goods model determined to be an electric appliances/computer series.
In the case of the goods model determined to be the electric appliances/computer series, since most goods model names are formed of a combination of alphabetic characters, numerals, and special characters, and most Korean words are just description of the goods, when performing the search according to the query matching method shown in FIGS. 1 and 2, a very suitable search result may be outputted.
Hereinafter, a query matching method according to another embodiment of the present invention will be described.
FIG. 6 is a flowchart illustrating a query matching method according to another embodiment of the present invention. The query matching method according to the present embodiment may be performed with respect to a predetermined database including an index for a search. Also, the query matching method may be performed by a predetermined query matching system.
In S601, the query matching system searches the database by using a first query generated by removing spaces or special characters from an original query, and generates a first search result.
For example, the query matching system may generate the first query by removing ' ', 7, '-', and '+' from the original query and may generate the first search result including the first query by searching the database. In S602, the query matching system determines whether a number of the first search results is less than a predetermined reference value.
For example, the predetermined reference value may be determined to be any one of 2 through 10. For example, when the predetermined reference value is 5, the query matching system may determine whether to perform a next operation by determining whether the number of the first search results is less than 5 in S602.
When the number of the first search results is determined to be not less than the predetermined reference value in S602, in S606, the first search result is outputted as a total search result. When the number of the first search results is determined to be less than the predetermined reference value in S602, S603 will be performed.
In S603, when the number of the first search results is less than the predetermined reference value, the query matching system divides the original query into a plurality of words according to a space, searches the database by using a word whose first and last characters are alphabetic characters or numerals and including at least one numeral as a second query, and generates a second search result.
For example, when the original query is "Tag Heuher sunglasses [1000002538]", the query matching system may perform search by determining '1000002538' as the second query in S603.
For example, when the original query is "[6 monthly installments without interest] Ferragamo bracelet [34-282 l_esp]", the query matching system may perform search by determining '34-2821_esp' as the second query.
On the other hand, the query matching system may arrange a plurality of the second queries in a descending order of length, may search the database according to the order, and may generate the second search result.
In 604, the query matching system determines whether a sum of the number of the first search results and a number of the second search results is less than the predetermined reference value. For example, when the reference value is 5, the query matching system may determine whether to perform a next operation by determining whether the sum is less than 5 in S604.
When the sum is determined to be not less than the predetermined reference value in S604, in S606, the first search results and the second search results are outputted as the total search result. When the sum is determined to be less than the predetermined reference value in S604, S605 will be performed.
In S605, the query matching system divides the second query into a plurality of tokens by special characters, searches the database by using a third query generated by connecting the tokens by space, and generates a third search results. For example, the query matching system may generate the tokens by dividing the second query by V, '-', '+', '_', and V.
Also, in S605, the query matching system may generate the third search result via a process of decreasing a number of the tokens included in the third query from a latter part of the third query while repeatedly searching the database.
For example, when the third query is "34-2821_esp", the query matching system may repeatedly search the database in an order of '34 2821 esp1, '34 2821', and '34'.
For example, when the third query is "DGB 020 BE", the query matching system may repeatedly search the database while decreasing the number of the tokens included in the third query from the latter part in an order of 1DGB 020 BE', 1DGB 020', and 1DGB'. Also, in this case, when it is determined that a search result exists while performing S605, the query matching system may suspend the repeated search process and may generate the existing search result as the third search result.
In S606, the query matching system outputs at least one of the first through third search results as the total search result matched with the original query. Also, the index for a search, included in the database, may include a plurality of search fields including a goods model name field and a keyword field, and S601 through S606 shown in FIG. 6 may be performed limited to the goods model name field and the keyword field of the index for a search. Also, though not shown in FIG. 6, S 106 through SI lO described with reference to FIG. 1 may be additionally performed in the query matching method according to the present embodiment.
On the other hand, the query matching method according to the present embodiment may be performed by a predetermined goods information registration system having the query matching system according to the present invention, the index for a search, included in the database, may include a plurality of search fields associated with a goods model, and the goods model may be included in a brand-name item series.
In the case of the goods model included in the brand-name item series, since a name of the goods model is generally formed of numerals or a combination of alphabetic characters and numerals and a Korean goods name is generally only a description, a very suitable search result may be outputted when performing the search according to the query matching method according to the present embodiment.
Hereinafter, a query matching method according to still another embodiment of the present invention will be described with reference to FIG. 6. The query matching method according to the present embodiment may be performed with respect to a predetermined database including an index for a search and may be performed by a predetermined query matching system.
In S601, the query matching system searches the database by using a first query generated by removing a space or a special character from an original query and generates a first search result. S602 is identical with the description of the previous embodiment.
In S603, the query matching system searches the database by using a word whose first and last characters are alphabetic characters or numerals and including at least one numeral as a second query and generates a second search result. S604 is identical with the description of the previous embodiment.
In S605, the query matching system searches the database by using a third query generated by morphologically analyzing the original query and generates a third search results. In S606, the query matching system outputs at least one of the first through third search results as the total search result matched with the original query.
On the other hand, the query matching method according to the present embodiment may be performed by a predetermined goods information registration system having the query matching system according to the present invention, the index for a search, included in the database, may include a plurality of search fields associated with a goods model, and the goods model may be included in cosmetics, powdered milk, diapers, or other general goods series.
In the case of cosmetics, powdered milk, diapers, or other general goods series, since a name of the goods model is generally in Korean, a very suitable search result may be outputted when performing the search according to the query matching method according to the present embodiment.
FIG. 7 is a diagram illustrating still another example of a search result view capable of being outputted by performing the query matching method according to the present invention. Since the conventional shopping mall relay server having a general database search system and search is not performed by a search engine, it is not easy to search by using a Korean goods name. Since names of cosmetics and powdered milk are generally in Korean, it is very difficult to output a search result according to a conventional technology.
However, according to the present invention, since a search associated with a
Korean goods name is performed by processing a goods name 701 corresponding to an original query according to various search logics of the present embodiment, a goods model matched with the goods name 701 may be precisely searched and outputted as shown in 702.
The query matching method according to the present invention may be embodied as a program instruction capable of being executed via various computer units and may be recorded in a computer-readable recording medium. The computer readable medium may include a program instruction, a data file, and a data structure, separately or cooperatively. The program instructions and the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those skilled in the art of computer software arts. Examples of the computer-readable media include magnetic media (e.g., hard disks, floppy disks, and magnetic tapes), optical media (e.g., CD-ROMs or DVD), magneto-optical media (e.g., optical disks), and hardware devices (e.g., ROMs, RAMs, or flash memories, etc.) that are specially configured to store and perform program instructions. The media may also be transmission media such as optical or metallic lines, wave guides, etc. including a carrier wave transmitting signals specifying the program instructions, data structures, etc. Examples of the program instructions include both machine code, such as produced by a compiler, and files containing high- level language codes that may be executed by the computer using an interpreter. The hardware elements above may be configured to act as one or more software modules for implementing the operations of this invention.
Hereinafter, a query matching system according to the present invention will be described.
FIG. 8 is a block diagram illustrating a configuration of a query matching system 800 according to an embodiment of the present invention. The query matching system 800 according to the present embodiment may perform a search process with respect to a predetermined database including an index for a search.
Referring to FIG. 8, the query matching system 800 according to the present embodiment includes a first search unit 801, a second search unit 802, a third search unit 803, and an output unit 804.
The first search unit 801 searches the database by using a first query generated by removing spaces or special characters from an original query and generates a first search result.
For example, the first search unit 801 may generate the first query by removing ' ', V, '-', and '+' from the original query and may generate the first search query including the first query by searching the database. When a number of the first search results is less than a predetermined reference value, the second search unit 802 divides the original query into a plurality of words according to a space, searches the database by using a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters from the plurality of words, as a second query, and generates a second search result.
For example, the second search unit 802 may divide the original query by ' ', generate words formed of only alphabetic characters, numerals, and '/', '-', and '+' as the second query, and search the database by using the second query, thereby generating the second search result. The described process is performed on the assumption that English words are more important. Words determined to be unnecessary for searching the database according to the assumption are removed. When a sum of the number of the first search results and a number of the second search results is less than the predetermined reference value, the third search unit 803 searches the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral as a third query and generates a third search result. In this case, the third search unit 803 may arrange a plurality of the third queries in a descending order of length and may search the database according to the order, and may generate the third search result.
In this case, the third search unit 803 may perform S201 through S208 shown in FIG. 2. The output unit 804 outputs at least one of the first search results, the second search results, and the third search results, as a total search result matched with the original query. On the other hand, the query matching system 800 according to the present embodiment may operate being included in a predetermined goods information registration system, an index for a search, included in the database, may include a plurality of search fields associated with a goods model, and the goods model may be included in an electric appliances/computer series. Since a name of the goods model included in the electric appliances/computer series is generally formed of a combination of alphabetic characters, numerals, and special characters and Korean is generally used for only a description with respect to the goods, a very suitable search result may be performed when performing the search by the query matching system 800 according to the present embodiment.
Hereinafter, a query matching system 800 according to another embodiment of the present invention will be described with reference to FIG. 8.
The query matching system 800 according to the present embodiment may perform a search process with respect to a predetermined database including an index for a search and includes a first search unit 801, a second search unit 802, a third search unit 803, and an output unit 804.
The first search unit 801 searches the database by using a first query generated by removing spaces or special characters from an original query and generates a first search result. When a number of the first search results is less than a predetermined reference value, the second search unit 802 divides the original query into a plurality of words according to a space, searches the database by using a word whose first and last characters are alphabetic characters or numerals and including at least one numeral as a second query, and generates a second search result. For example, when the original query is "Tag Heuher sunglasses
[1000002538]", the second search unit 802 may perform search by determining '1000002538' as the second query.
For example, when the original query is "[6 monthly installments without interest] Ferragamo bracelet [34-282 l_esp]", the second search unit 802 may perform search by determining ' 34-282 l_esp' as the second query.
When a sum of the number of the first search results and a number of the second search results is less than the predetermined reference value, the third search unit 803 divides the second query into a plurality of tokens by special characters, searches the database by using a third query generated by connecting the tokens by a space, and generates a third search results.
For example, the third search unit 803 may generate the tokens by dividing the second query by '/', '-', '+', '_', and '.'. Also, the third search unit 803 may generate the third search result via a process of decreasing a number of the tokens included in the third query from a latter part of the third query while repeatedly searching the database.
For example, when the third query is "34-282 l_esp", the third search unit 803 may repeatedly search the database in an order of '34 2821 esp1, '34 2821', and '34'. The output unit 804 outputs at least one of the first search results, the second search results, and the third search results as a total search result matched with the original query.
On the other hand, the query matching system 800 according to the present embodiment may operate being included in a predetermined goods information registration system, an index for a search, included in the database, may include a plurality of search fields associated with a goods model, and the goods model may be included in a brand-name item series. Since a name of the goods model included in the brand-name item is generally formed of numerals or a combination of alphabetic characters and numerals, and Korean is generally used for only a description with respect to the goods, a very suitable search result may be performed when performing the search by the query matching system 800 according to the present embodiment.
Hereinafter, a query matching system 800 according to still another embodiment of the present invention will be described with reference to FIG. 8.
The query matching system 800 according to the present embodiment may perform a search process with respect to a predetermined database including an index for a search and includes a first search unit 801, a second search unit 802, a third search unit 803, and an output unit 804.
The first search unit 801 searches the database by using a first query generated by removing spaces or special characters from an original query and generates a first search result.
When a number of the first search results is less than a predetermined reference value, the second search unit 802 searches the database by using a second query generated by dividing the original query into a plurality of words according to a space and generates a second search result.
When a sum of the number of the first search results and a number of the second search results is less than the predetermined reference value, the third search unit 803 searches the database by using a third query generated by morphologically analyzing the original query and generates a third search result.
The output unit 804 outputs at least one of the first search results, the second search results, and the third search results as a total search result matched with the original query. On the other hand, the query matching system 800 according to the present embodiment may operate being included in a predetermined goods information registration system, an index for a search, included in the database, may include a plurality of search fields associated with a goods model, and the goods model may be included in cosmetics, powdered milk, diapers, or other general goods series. Since a name of the goods model included in the cosmetics, powdered milk, diapers, or other general goods series is generally in Korean, a more suitable search result may be performed when performing the search by the query matching system 800 according to the present embodiment, than the previous embodiments.
FIG. 9 is a block diagram illustrating a configuration of a query matching system 900 according to yet another embodiment of the present invention.
The query matching system 900 according to the present embodiment may perform a search process with respect to a predetermined database including an index for a search and includes a division unit 901, an extraction unit 902, a determination unit 903, a search unit 904, and an output unit 905 from elements shown in FIG. 9. The division unit 901 divides an original query into a plurality of words according to a space.
The determination unit 902 determines a word formed of any one of or a composite of two or more of alphabetic characters, numerals, and special characters from the plurality of words, as a search query. The search unit 904 searches the database by using the search query and generates a search result.
The output unit 905 outputs the search result as a search result matched with the original query.
Hereinafter, a query matching system 900 according to another embodiment of the present invention will be described with reference to FIG. 9.
The query matching system 900 according to the present embodiment may perform a search process with respect to a predetermined database including an index for a search and include a division unit 901, an extraction unit 902, a determination unit 903, a search unit 904, and an output unit 905 shown in FIG. 9.
The division unit 901 divides an original query into a plurality of words according to a space. The extraction unit 902 extracts a word formed of any one of or a composite of two or more of alphabetic characters, numerals, and special characters, from the plurality of words.
The determination unit 903 determines a word whose first and last characters are alphabetic characters or numerals and including at least one numeral, from the extracted words, as a search query.
The search unit 904 searches the database by using the search query and generates a search result.
The output unit 905 outputs the search result as a search result matched with the original query. Hereinafter, a query matching system 900 according to still another embodiment of the present invention will be described with reference to FIG. 9.
The query matching system 900 according tot the present embodiment may perform a search process with respect to a predetermined database including an index for a search and include a token generation unit 910 including a division unit 901 and an extraction unit 902, a determination unit 903, a search unit 904, and an output unit 905 shown in FIG. 9.
The token generation unit 910 generates a plurality of tokens by processing an original query according to predetermined division logic.
The token generation unit 910 may include the division unit 901 dividing the original query into a plurality of words according to a space and the extraction unit 902 extracting a word formed of any one of or a composite of two or more of alphabetic characters, numerals, and special characters, from the plurality of words, and may generate the plurality of tokens by dividing the extracted word into the plurality of tokens formed of a group of only alphabetic characters or only numerals or dividing the extracted word by special characters.
The determination unit 903 determines a search query by connecting the plurality of tokens by a space.
The search unit 904 repeatedly searches the database while decreasing a number of the tokens included in the search query from a latter part of the search query and generates a search result.
The output unit 905 outputs the search result as a search result matched with the original query.
Although a few embodiments of the present invention have been shown and described, the present invention is not limited to the described embodiments. Instead, it would be appreciated by those skilled in the art that changes may be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
Industrial Applicability
According to the present invention, there is provided a query matching method and system which can overcome a conventional database search method limitation of only providing a search result precisely matching an original query and can output a predetermined number of search results capable of being matched with the original query by processing the original query via a plurality of processes according to various search logics.
According to the present invention, there is also provided a query matching method and system in which an automated goods registration/match operation is embodied while registering goods information registered in a shopping mall in a shopping mall relay server, thereby notably reducing associated operation resource.
According to the present invention, there is also provided a query matching method and system in which different search logic is applied according to a property of goods corresponding to a respective query, thereby efficiently retrieving a goods model having high probability of being matched with a corresponding query.
According to the present invention, there is also provided a query matching method and system in which a suitable query is extracted from an original query that is a goods name inputted to a shopping mall server by a goods registrant and a goods model matched with a corresponding goods is retrieved by using the extracted query, thereby pre-performing goods registration/match operation when the goods registrant registers a goods in a shopping mall server, without additional manual operation for the goods registration/match when registering the goods in a shopping mall relay server.

Claims

1. A method of matching a query with respect to a predetermined database including an index for a search, the method comprising: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; dividing the original query into a plurality of words according to space, searching the database by using a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters from the plurality of words, as a second query, and generating a second search result, when a number of the first search results is less than a predetermined reference value; searching the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, selected from the second queries, as a third query and generating a third search result, when a sum of the first and second search results is less than the predetermined reference value; and outputting at least one of the first search result, the second search result, and the third search result, as a total search result matched with the original query.
2. The method of claim 1, further comprising separating the original query into tokens by using a predetermined search engine indexer, searching the database by using the tokens, and generating a fourth search result, when a sum of the first through third search results is less than the predetermined reference value, wherein the total search result comprises the fourth search result.
3. The method of claim 2, further comprising generating a fifth search result including at least one of the tokens from the database when a sum of the first through fourth search results is less than the predetermined reference value, wherein the total search result comprises the fifth search result.
4. The method of claim 1, wherein: the index for a search comprises a plurality of search fields comprising a goods model name field and a keyword field; and the searching the database by a third query and generating a third search result is performed limited to the goods model name field and the keyword field of the index for a search.
5. The method of claim 4, wherein the searching the database by a third query and generating a third search result comprises: arranging a plurality of the third queries in a descending order of length; and searching the database according to the arranged order and generating the third search result.
6. The method of claim 1 , wherein the searching the database by a third query and generating a third search result comprises: separating the third query into tokens formed of a group of only alphabetic characters or only numerals ; generating the fourth query by connecting the tokens by space; and searching the database by using the fourth query and generating the third search result.
7. The method of claim 6, wherein the searching the database by using the fourth query and generating the third search result comprises generating the third search result via a process of repeatedly searching the database while decreasing a number of the tokens included in the fourth query from a latter part of the fourth query.
8. The method of claim 7, wherein, when a search result is generated while performing the process, the searching the database by using the fourth query and generating the third search result comprises generating the search result as the third search result and suspending the process.
9. The method of claim 7, wherein, when the tokens that remain after performing the process of searching the database are corresponding to the token formed of the group of only alphabetic characters, the searching the database by using the fourth query and generating the third search result comprises searching the database by removing a space between the left tokens.
10. The method of claim 1 , wherein: the method of matching a query is performed by a predetermined goods information registration system; the index for a search comprises a plurality of search fields associated with a goods model; and the goods model is classified as a electric home appliances series.
11. A method of matching a query with respect to a predetermined database including an index for a search, the method comprising: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; dividing the original query into a plurality of words according to a space, searching the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, from the plurality of words, as a second query, and generating a second search result, when a number of the first search results is less than a predetermined reference value; generating a plurality of tokens by dividing the second query into special characters, searching the database by using a third query generated by connecting the tokens by a space, and generating a third search result, when a sum of the first and second search results is less than the predetermined reference value; and outputting at least one of the first search result, the second search result, and the third search result, as a total search result matched with the original query.
12. The method of claim 11, wherein the searching the database by using a third query and generating a third search result comprises generating the third search result via a process of repeatedly searching the database while decreasing a number of the tokens included in the third query from a latter part of the third query.
13. The method of claim 11 , wherein: the method of matching a query is performed by a predetermined goods information registration system; the index for a search comprises a plurality of search fields associated with a goods model; and the goods model is classified into a brand-name item series.
14. A method of matching a query with respect to a predetermined database including an index for a search, the method comprising: searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; searching the database by using a second query generated by dividing the original query according to a space and generating a second search result, when a number of the first search results is less than a predetermined reference value; and outputting the first search result or the second search result as a total search result matched with the original query.
15. The method of claim 14, further comprising, when a sum of the first search results and the second search results is less than the predetermined reference value, searching the database by using a third query that is generated by performing morphological analysis on the original query, and generating a third search result, wherein the total search result comprises the third search result.
16. The method of claim 14, wherein: the method of matching a query is performed by a predetermined goods information registration system; the index for a search comprises a plurality of search fields associated with a goods model; and the goods model is classified into cosmetics, powdered milk, diapers, or other general goods type.
17. A method of matching a query with respect to a predetermined database including an index for a search, the method comprising: dividing an original query into a plurality of words according to a space; determining a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters, from the plurality of words, as a search query; searching the database by using the search query and generating a search result; and outputting the search result as a search result matched with the original query.
18. A method of matching a query with respect to a predetermined database including an index for a search, the method comprising: dividing an original query into a plurality of words according to a space; extracting a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters, from the plurality of words; determining a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral from the extracted words as a search query; searching the database by using the search query and generating a search result; and outputting the search result as a search result matched with the original query.
19. A method of matching a query with respect to a predetermined database including an index for a search, the method comprising: generating a plurality of tokens by processing an original query according to predetermined separation logic; determining a search query by connecting the plurality of tokens by a space; repeatedly searching the database while decreasing a number of the tokens included in the search query from a latter part of the search query and generating a search result; and outputting the search result as a search result matched with the original query.
20. The method of claim 19, wherein the generating a plurality of tokens by processing an original query according to predetermined separation logic comprises: dividing the original query into a plurality of words according to space; extracting a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters, from the plurality of words; and generating the plurality of tokens by dividing the extracted word into the plurality of tokens formed of a group of only the alphabetic characters or only the numeral or dividing the extracted into the special characters.
21. A computer-readable recording medium in which a program for executing the method according to any one of claims 1 through 20 is recorded.
22. A query matching system performing query matching with respect to a predetermined database including an index for a search, the system comprising: a first search unit searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; a second search unit dividing the original query into a plurality of words according to a space, searching the database by using a word formed of any one or a composite of at least two of alphabetic characters, numerals, and special characters, from the plurality of words, as a second query, and generating a second search result, when a number of the first search results is less than a predetermined reference value; a third search unit searching the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral, selected from the second queries, as a third query and generating a third search result, when a sum of the first and second search results is less than the predetermined reference value; and an output unit outputting at least one of the first search result, the second search result, and the third search result, as a total search result matched with the original query.
23. A query matching system performing query matching with respect to a predetermined database including an index for a search, the system comprising: a first search unit searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; a second search unit dividing the original query into a plurality of words according to a space, searching the database by using a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one number, from the plurality of words, as a second query, and generating a second search result, when a number of the first search results is less than a predetermined reference value; a third search unit generating a plurality of tokens by dividing the second query into special characters, searching the database by using a third query generated by connecting the tokens by a space, and generating a third search result, when a sum of the first and second search results is less than the predetermined reference value; and an output unit outputting at least one of the first search result, the second search result, and the third search result, as a total search result matched with the original query.
24. A query matching system performing query matching with respect to a predetermined database including an index for a search, the system comprising: a first search unit searching the database by using a first query generated by removing a space or a special character from an original query and generating a first search result; a second search unit searching the database by using a second query generated by dividing the original query according to a space and generating a second search result, when a number of the first search results is less than a predetermined reference value; and an output unit outputting the first search result or the second search result as a total search result matched with the original query.
25. A query matching system performing query matching with respect to a predetermined database including an index for a search, the system comprising: a division unit dividing an original query into a plurality of words according to a space; a determination unit determining a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters, from the plurality of words, as a search query; a search unit searching the database by using the search query and generating a search result; and an output unit outputting the search result as a search result matched with the original query.
26. A query matching system performing query matching with respect to a predetermined database including an index for a search, the system comprising: a division unit dividing an original query into a plurality of words according to a space; an extraction unit extracting a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters, from the plurality of words; a determination unit determining a word whose first character is an alphabetic character and last character is an alphabetic character or a numeral and including at least one numeral from the extracted words as a search query; a search unit searching the database by using the search query and generating a search result; and an output unit outputting the search result as a search result matched with the original query.
27. A query matching system performing query matching with respect to a predetermined database including an index for a search, the system comprising: a token generation unit generating a plurality of tokens by processing an original query according to predetermined separation logic; a determination unit determining a search query by connecting the plurality of tokens by a space; a search unit repeatedly searching the database while decreasing a number of the tokens included in the search query from a latter part of the search query and generating a search result; and an output unit outputting the search result as a search result matched with the original query.
28. The system of claim 27, wherein the token generation unit comprises: a division unit dividing the original query into a plurality of words according to space; and an extraction unit extracting a word formed of any one of or a composite of at least two of alphabetic characters, numerals, and special characters, from the plurality of words; and generates the plurality of tokens by dividing the extracted word into the plurality of tokens formed of a group of only alphabetic characters or only numerals or dividing the extracted word into special characters.
PCT/KR2006/001890 2005-05-20 2006-05-19 Query matching system and method, and computer readable recording medium recording program for implementing the method WO2006123918A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2008512220A JP5027803B2 (en) 2005-05-20 2006-05-19 Query matching system and method, and computer-readable recording medium on which a program for executing the method is recorded

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
KR1020050042470A KR100623628B1 (en) 2005-05-20 2005-05-20 Query matching system and method, and computer readable recording medium recording program for implementing the method
KR10-2005-0042470 2005-05-20
KR1020050042481A KR100709984B1 (en) 2005-05-20 2005-05-20 Query matching method and system for outputting results matched to query by processing the query according to various logics
KR10-2005-0042481 2005-05-20

Publications (1)

Publication Number Publication Date
WO2006123918A1 true WO2006123918A1 (en) 2006-11-23

Family

ID=37431462

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2006/001890 WO2006123918A1 (en) 2005-05-20 2006-05-19 Query matching system and method, and computer readable recording medium recording program for implementing the method

Country Status (2)

Country Link
JP (1) JP5027803B2 (en)
WO (1) WO2006123918A1 (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9092517B2 (en) 2008-09-23 2015-07-28 Microsoft Technology Licensing, Llc Generating synonyms based on query log data
US9229924B2 (en) 2012-08-24 2016-01-05 Microsoft Technology Licensing, Llc Word detection and domain dictionary recommendation
US20160034532A1 (en) * 2014-07-31 2016-02-04 Dmytro Andriyovich Ivchenko Flexible operators for search queries
US9594831B2 (en) 2012-06-22 2017-03-14 Microsoft Technology Licensing, Llc Targeted disambiguation of named entities
US10032131B2 (en) 2012-06-20 2018-07-24 Microsoft Technology Licensing, Llc Data services for enterprises leveraging search system data assets

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103488648B (en) 2012-06-13 2018-03-20 阿里巴巴集团控股有限公司 A kind of multilingual mixed index method and system
JPWO2014196063A1 (en) * 2013-06-06 2017-02-23 株式会社野村総合研究所 Product search system and product search program
JP6102979B2 (en) * 2015-05-12 2017-03-29 株式会社Jvcケンウッド Information selection device, information selection method, and computer program
JP2017102977A (en) * 2017-03-06 2017-06-08 株式会社野村総合研究所 Product retrieval system and product retrieval program

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR19990031784A (en) * 1997-10-14 1999-05-06 정선종 Natural Language Query Processing Method for E-Commerce Product Information Retrieval
US5920854A (en) * 1996-08-14 1999-07-06 Infoseek Corporation Real-time document collection search engine with phrase indexing
KR20020045971A (en) * 2000-12-12 2002-06-20 오길록 Method for product detailed information extraction of internet shopping mall with ontology and wrapper data
US6748387B2 (en) * 1998-06-30 2004-06-08 Microsoft Corporation Flexible keyword searching
US6751606B1 (en) * 1998-12-23 2004-06-15 Microsoft Corporation System for enhancing a query interface

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06259480A (en) * 1993-03-10 1994-09-16 Hitachi Ltd Document retrieving method
JPH09198450A (en) * 1996-01-17 1997-07-31 Sharp Corp Method for managing address data including postal code
JPH1011431A (en) * 1996-06-19 1998-01-16 Okinawa Nippon Denki Software Kk Kanji retrieval device and method
US5982370A (en) * 1997-07-18 1999-11-09 International Business Machines Corporation Highlighting tool for search specification in a user interface of a computer system
JP2002269116A (en) * 2001-03-13 2002-09-20 Ricoh Co Ltd System and program for retrieving document
JP4185681B2 (en) * 2001-07-30 2008-11-26 株式会社みずほ銀行 Transfer processing method
JP3898016B2 (en) * 2001-09-28 2007-03-28 三菱電機株式会社 Information search device, information search method, and information search program
JP2004192344A (en) * 2002-12-11 2004-07-08 Sharp Corp Information processor
JP2004246422A (en) * 2003-02-10 2004-09-02 Hitachi Software Eng Co Ltd Information retrieval support device
JP2004326293A (en) * 2003-04-23 2004-11-18 Canon Inc Document retrieving system
JP2005055782A (en) * 2003-08-07 2005-03-03 Sharp Corp Data input system, handy terminal, data input method, program, and recording medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5920854A (en) * 1996-08-14 1999-07-06 Infoseek Corporation Real-time document collection search engine with phrase indexing
KR19990031784A (en) * 1997-10-14 1999-05-06 정선종 Natural Language Query Processing Method for E-Commerce Product Information Retrieval
US6748387B2 (en) * 1998-06-30 2004-06-08 Microsoft Corporation Flexible keyword searching
US6751606B1 (en) * 1998-12-23 2004-06-15 Microsoft Corporation System for enhancing a query interface
KR20020045971A (en) * 2000-12-12 2002-06-20 오길록 Method for product detailed information extraction of internet shopping mall with ontology and wrapper data

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9092517B2 (en) 2008-09-23 2015-07-28 Microsoft Technology Licensing, Llc Generating synonyms based on query log data
US10032131B2 (en) 2012-06-20 2018-07-24 Microsoft Technology Licensing, Llc Data services for enterprises leveraging search system data assets
US9594831B2 (en) 2012-06-22 2017-03-14 Microsoft Technology Licensing, Llc Targeted disambiguation of named entities
US9229924B2 (en) 2012-08-24 2016-01-05 Microsoft Technology Licensing, Llc Word detection and domain dictionary recommendation
US20160034532A1 (en) * 2014-07-31 2016-02-04 Dmytro Andriyovich Ivchenko Flexible operators for search queries

Also Published As

Publication number Publication date
JP5027803B2 (en) 2012-09-19
JP2008546051A (en) 2008-12-18

Similar Documents

Publication Publication Date Title
KR100721406B1 (en) Product searching system and method using search logic according to each category
WO2006123918A1 (en) Query matching system and method, and computer readable recording medium recording program for implementing the method
US8037064B2 (en) Method and system of selecting landing page for keyword advertisement
US10467265B2 (en) Method for extracting entries from a database
JP4750814B2 (en) Advertising method and system for exposing contextual advertising information
US7702700B2 (en) Automatic matching method and automatic matching system
US20090240670A1 (en) Uniform resource identifier alignment
CN106547794B (en) Information searching method and device
US20130275451A1 (en) Systems And Methods For Contract Assurance
WO2020248377A1 (en) Information pushing method and apparatus, computer readable storage medium, and computer device
Melli Shallow semantic parsing of product offering titles (for better automatic hyperlink insertion)
KR100709984B1 (en) Query matching method and system for outputting results matched to query by processing the query according to various logics
WO2021055868A1 (en) Associating user-provided content items to interest nodes
CN112749325A (en) Training method and device for search ranking model, electronic equipment and computer medium
CN109902215A (en) A kind of method and system of deals match
KR100769465B1 (en) Query matching method and system using category matching
CN109101630A (en) A kind of generation method, device and the equipment of application program search result
CN104376034A (en) Information processing apparatus, information processing method, and program
KR101105947B1 (en) Commodity information registering method and system which automatically matches commodity model with the commodity information
KR101137056B1 (en) Commodity information registering method and system which uses automatic commodity model suggesting logic
CN109165297B (en) Universal entity linking device and method
CN110245215B (en) Text retrieval method and device
KR100623628B1 (en) Query matching system and method, and computer readable recording medium recording program for implementing the method
Jalal et al. A web content mining application for detecting relevant pages using Jaccard similarity
Schmidts et al. Catalog Integration of Low-quality Product Data by Attribute Label Ranking.

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application
ENP Entry into the national phase

Ref document number: 2008512220

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

NENP Non-entry into the national phase

Ref country code: RU

122 Ep: pct application non-entry in european phase

Ref document number: 06768546

Country of ref document: EP

Kind code of ref document: A1