Information service system and method for supplying information
The present invention concerns an information service system which supplies requested information to a customer by means of a data communications network. The present invention also concerns a method for supplying information to a cus- tomer.
Accessing of information has always been important in the activities of companies and individuals. Information is available not only in verbal form and written on paper, but also in electric form. The significance of the latter has increased with the development of computer and network technology. In conventional computer-based information services, information is searched from suitable databases by means of search words. The disadvantage of these is that a search usually produces only titles and abstracts of reference publications, whereupon the publication itself must be acquired from elsewhere. Furthermore, the most beneficial field-specific databases are not open; using them causes fixed costs in addition to search costs. Another dis- advantage is that efficient information search requires professional skill, in practice often the assistance of an informatician.
The expansion of the Internet has made an enormous amount of information available to nearly everyone. The result of an information search performed on the Internet is, however, quite random and the quality of the information is inconsistent. The most valuable information is still usually contained in chargeable databases. Another disadvantage of the Internet as well as other known information search methods is that information is only available in a certain format: the terminal used by the person searching for information must be compatible with said format. Furthermore, the information is most often limited to a text format or a text and image format. If sound or motion pictures are included, they are chronologically independent from possible text.
A system is known from publication JP 7105113 in which a server belonging to a local area network functions as an adapter for image information. When an image file addressed to a workstation arrives from outside, the server finds out whether the file is in a format understood by the workstation. If not, the server performs a necessary conversion. The system does not perform actual information search. Moreover, the amount of media types the system can handle are quite limited.
The purpose of the invention is to reduce the mentioned disadvantages of the prior art. A system according to the invention is characterized by what is presented in
independent claim 1. A method according to the invention is characterized by what is presented in independent claim 6. Advantageous embodiments of the invention are presented in the dependent claims.
The basic idea of the invention is as follows: An information server can be con- nected to numerous databases which function as sources. Information retrieved from a source is converted by the server to an internal format used by the server. Furthermore, the information is processed and arranged to produce chronologically parallel text, image and sound signals. Being parallel means the signal indicates to which part of the text each image and sound message is related. The server receives information requests from customers. If information corresponding to a request does not happen to exist in the server's own database, the server searches for the information and arranges it in the above-mentioned manner. Then the server further arranges the retrieved information, which is in an internal format into a format(s) appearing in the information request and merges the result into an information packet to be sent to the customer. In a normal case said packet includes text, images and sound.
An advantage of the invention is that relevant information in a desired format can be supplied to a customer, i.e., an information user. This comes from the automatic request handling and format conversion functions of the server. Another advantage of the invention is that the media type of the information retrieved from a source can be automatically modified. For example, a corresponding speech signal can be produced parallel to a text. Still another advantage of the invention is that retrieved information can be refined by selecting only essential parts of it. Yet another advantage of the invention is that a customer gets information selected from a large num- ber of databases and net pages using relatively simple requests.
The invention is described in detail below, with reference to the enclosed drawings, in which
figure 1 presents an example of a system according to the invention,
figure 2 presents an example of an information conversion unit according to the invention,
figure 3 presents another example of an information conversion unit according to the invention and
figure 4 presents as a flow chart an example of a method according to the invention.
Figure 1 is an example of an information service system according to the invention. The central part of the system is an information server 100. The server comprises an information request handling unit 110, a conversion unit 120, a media adaptor 130, a media merger 140 and a network interface unit 150. The latter contains an interface to at least one data communications network in order to establish a connection to information sources and customers. Data communications networks may be various data networks and also telephone and mobile communications networks. Figure 1 shows, as an example, a first information source SI, a second information source S2 and another information source Sn interfaced to data communications networks. Said sources may be, for example, field-specific databases or hypertext pages accessible through the Internet. The server 100 has access to a representative group of chargeable information sources, too. The figure has a dotted line representing a connection SC to the second information source S2. Another dotted line represents a connection UC to a terminal UT. A user, i.e., a customer may with the terminal use the information service according to the invention.
The interface unit 150 directs messages coming from customers to the information request handling unit 110. This analyzes each message, and if the message is inter- preted as an information request, informs the conversion unit 120 about the information request by means of an internal message. The request handling unit 110 is also connected to the media adaptor 130 for informing the adaptor about data formats used by a customer, which formats have appeared in the customer's information request.
The task of the conversion unit 120 is to process information retrieved from external information sources. The information is converted to a format used only inside the server. In this description the format of the server's internal information is referred to as RDF (rich data format). Conversion from one format to another takes place via the server's internal format RDF. This reduces the total amount of conversion code needed when there are several formats in a certain media.
The task of the media adaptor 130 is to further convert the data in RDF mode received from the conversion unit into the format(s) appearing in the customer's information request. Thus it is realized the above-mentioned conversion from an original format to another format used by the customer. The media merger 140 re- ceives data from the media adaptor. It combines data in different media types into
an information packet corresponding to the customer's request. That information packet is transferred to the network interface unit 150, which sends the packet to the customer.
In the example of figure 1, the information server 100 also has its own database 160. The server sets up this database from information retrieved from external sources. The criterion for storing information is its statistical occurrence as an object of customer requests. Naturally, the content of the database 160 is in RDF mode. The structure of the server 100 in figure 1 is to be understood as functional. What kind of physical units and buses there are in the server can be designed freely.
Figure 2 shows an example of an information conversion unit according to the invention. As functional parts the conversion unit 220 comprises a format conversion layer 221, a media conversion layer 222 and a media linking layer 223. The format conversion layer contains modules, each of which makes conversions- within a media type from an original format to the server's internal format. In figure 2 the se- lected media types are text, audio, image and video. The text module converts e.g. a Word document into server's internal text format. The latter is referred to as RTF (rich text format). The audio module converts MP3-packed sound, for example, into server's internal sound format. The image module converts e.g. an image made by Designer drawing application into server's internal image format. The video module converts e.g. MPEG4-coded (motion picture experts group) video signals into server's internal video format.
The media conversion layer 222 functions chronologically after the format conversion layer has completed its own processing. The media conversion layer contains modules, each of which makes a conversion from one media type to another. The following conversions were selected for the example of figure 2:
- Converting a text to a speech signal. This happens by means of a speech synthesizer.
- Converting a speech signal to text. This happens by means of a conversion program made for the server, which program provides text in RTF mode from a speech signal in the system's internal sound format.
- Extracting text from image files.
- Converting a video signal to a group of separate images.
The media linking layer 223 functions after the media conversion layer has completed its own processing. The linking layer couples signals of different media types together chronologically. Media types in figure 2 are text, sound and images. Each
of the three signals is stored on its own figurative information track, according to figure 2 the text signal on text track TR1, the sound signal on sound track TR2 and the image signal on image track TR3. The information tracks are parallel, so that the system knows to which text section a sound signal or image is related.
Figure 3 shows another example of an information conversion unit according to the invention. The conversion unit 320 contains a format conversion layer 321, a media conversion layer 322 and a media linking layer 323 as does the conversion unit of figure 2. The difference is that now the part storing RDF data has in addition to text, sound and image tracks also a fourth information track TR4, called a meta track. On the meta track there is stored data found in retrieved information and belonging to no one of the three other media types but linked to the content of other tracks. Such data could be, in the case of an image, coordinates of the recording point and the direction of light in relation to the recorded object.
Figure 4 presents as a flow chart an example of a method according to the inven- tion. The method describes the manner of operation of an information server. Operation begins with a request for information sent from a customer's terminal. The request is received by the server in phase 401. The request for information is analyzed in phase 402, and as a result necessary information is sent to the server's conversion unit and media adaptor. Next, in phase 403, the conversion unit finds out whether the requested information exists in the server's own database. If not, phase 404 follows, where based on a call sent by the conversion unit, a connection is made through a network to an information source selected by the conversion unit. In phase 405 information is received from the information source on the basis of search message sent to the information source. Naturally, search can be done from several information sources. Format conversions, media conversions and media linkings, described in the description of figure 2, are performed to the received information in phases 406, 407 and 408. As a result, RDF data is provided, which data is stored on media type-specific information tracks. In phase 409 a media adapter converts the RDF data into a formats suitable for a customer's terminal. If the request was sent using a WWW (world wide web) browser, for example, the media adapter converts the content of the text track to HTML (hyper text markup language) code and the content of the image track to GIF (graphics interchange format) code. If in turn the request was sent using a WAP (wireless application protocol) phone, for example, the media adapter converts the content of the text track to WML (wireless markup language) code. Next (phase 410) a media merger combines such an information packet from the data of different information tracks that
the customer's terminal outputs it in a descriptive form. If the request was sent by means of a WWW browser as in the above-mentioned example, the media merger inserts GIF code among the HTML code. If the system comprises a meta track mentioned in the description of figure 3, a three-dimensional model of the image object may be constructed, for example, based on the contents of the image and meta tracks. In the last phase 411 an interface unit sends the information packet to the customer's terminal.
If in phase 403, according to the criteria used by the database software, a sufficient amount of requested information exists in the server's own database, the process jumps directly to phase 409.
Certain embodiments of the invention are described above. The invention is not limited just to those embodiments. The inventive idea can be applied in different ways to the extent allowed by the independent claims 1 and 6.