JP2011186692A - Information retrieval system and information retrieval method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information retrieval system and an information retrieval method immediately acquiring relevant information from a network without separately starting a retrieval engine in a process of retrieving document information stored in the network. <P>SOLUTION: An analysis server 140 acquires data displayed on a browser 121 through the network 110 in response to a request from a client 120, extracts a word from document information included in the acquired data by using a morphological analysis means 141, finds out relevant document information related to the extracted word from a storage server 130 on the network 110, creates processed data by adding the existence information of the found out relevant document information to the acquired data, and causes the browser 121 of the client 120 to display the processed data through the network 110. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an information search system and an information search method for searching for information existing on a network, and more particularly to an information search system and an information search method suitable for searching for document information existing on the Internet.

  A search engine is used to find out required information from a lot of information existing on the network. The basic procedure for information retrieval using a search engine is as follows. That is, when a search condition is given from the client side, the search condition is sent to the server side via the network, and the search engine on the server side detects information that matches the search condition from the network, and the location of the information. Is returned from the server side to the client side via the network. A text search for searching only document information including a specific word or character string (text) from an enormous amount of data such as content existing on the Internet is an example of an information search according to the above procedure.

  Text search for document information distributed on the Internet is a database that actually stores data including document information, a search server for providing information search services to clients, and information search Clients for using the service can be used by being connected to the Internet via communication paths. A client is often composed of a personal computer (hereinafter abbreviated as a personal computer) equipped with a browser, and a user who wants to find desired document information on the Internet is a search portal site displayed on the browser of the personal computer. Usually, a character string (key character string) is input as search key information.

  The search result returned from the search server to the client may be document information itself, but is often location information of the document information on the Internet. In general, URLs (Uniform Resource Locators), which are link destination information of a plurality of document information that are candidates for search results, are displayed in a list format on a browser of a client. According to this display method, the user can access the document information and confirm the contents only by selecting each candidate URL displayed in the browser. If the user determines that the desired search result has been obtained, the user ends the search. Otherwise, the search procedure is repeated using a different key character string.

  By the way, in the above-described search process, not only target document information but also other information may be required. For example, when a text search is performed using a certain key character string, it is necessary to newly investigate an unknown term included in document information as a search result. Of course, it is possible to collect information on the term from the network by separately performing a new search using the unknown term to be newly examined as a key character string. However, it is very troublesome for a user who has already started a search to start a search engine separately and perform a new search in parallel.

  Therefore, conventionally, location information of related information is added in advance to words included in document information that can be searched on a network. More specifically, link destination information of related information is added in advance for main terms included in document information accumulated on the network (a so-called link is established). . According to this method, even if an unknown word is included in the document information obtained as a search result, it is selected by a method such as clicking on link destination information such as URL attached to the character string on the browser. It is possible to access related information existing at the link destination without performing a new search separately.

However, this method has a problem that it is difficult to link all document information that can be searched on the network without omission. In addition, there is a disadvantage that information on the network is changed and related information does not exist at the link destination (so-called broken link). The latter is particularly problematic for document information contained in Internet content that is constantly being updated, but it is practically impossible to keep all link information up to date on the Internet. .

  Therefore, as a method for avoiding the above problems, a word table created by cutting out words included in document information (document file) accumulated on the network is added to the original document file. Has been proposed (Patent Document 1). According to this proposal, the word table is automatically created by processing the document information in a timely manner. Specifically, when the created document information is accumulated, at night, etc. It is assumed that the system is created using a technique such as morphological analysis during a time period when the system is not used.

  However, even with the above method, as the amount of document information accumulated on the network increases, the time required for processing increases dramatically. Therefore, the word table is added to all document information during the time when the system is not used. It is clear that it becomes difficult to add without leakage. For the same reason, it is not guaranteed that the word table will be updated immediately as the document information is updated. Therefore, even if the above method is applicable to the document information stored in the document management system in the company built on the intranet, the document information on the Internet is continuously updated. It must be said that it cannot be applied at all.

JP 2000-29906 A

  The present invention has been made in view of such circumstances, and in the process of searching for document information stored on the network, related information is immediately acquired from the network without starting a search engine separately. It is an object to provide an information search system and an information search method that are possible.

  In order to solve the above problems, an information search system according to the present invention includes a storage server that stores data including document information, and a browser that displays the data including the document information, all connected to a network. An information retrieval system comprising: a client comprising: an analysis server comprising morphological analysis means; wherein the browser performs morphological analysis on document information contained in data displayed on the display means Analysis request means for requesting the analysis server, and in response to a request from the analysis request means, the analysis server acquires data displayed on the display means, and document information included in the acquired data A word is cut out from the morphological analysis from the storage server, the related document information related to the word is searched from the storage server, The location information of the writing information is added to the obtained data to create the processed data, the client displays the processed data to the browser, it is characterized in.

  In addition, another information retrieval system of the present invention for solving the above-described problems includes a storage server that is connected to a network and stores data including document information, and data including the document information. An information search system including a client that includes a browser that displays a message and an analysis server that includes a morphological analysis unit, wherein the browser is configured to process document information included in data displayed on the display unit. Analysis request means for requesting the analysis server to perform morphological analysis, and in response to a request from the analysis request means, the analysis server acquires the data displayed on the display means, and the acquired data A word is cut out from included document information by morphological analysis, and related document information related to the word is represented as part-of-speech information and attribute of the word. Searching from the storage server based on information, adding location information of the related document information to the acquired data to create processed data, and displaying the processed data on the browser by the client. It is a feature.

  In addition, the information search method of the present invention for solving the above-described problem is a storage server that is connected to a network and stores data including document information, and displays the data including the document information. An information search method used on a network system including a client having a browser and an analysis server having morphological analysis means, the morpheme for document information included in data displayed on the display means by the browser The analysis server includes a step of requesting the analysis server to execute analysis, the analysis server acquiring data displayed on the display unit in response to a request from the analysis request unit, and included in the acquired data Extracting words from document information by morphological analysis; and storing related document information related to the words. A step of searching from a server, a step of creating processed data by adding location information of the related document information to the acquired data, and a step of displaying the processed data on the browser by the client. It is characterized by that.

  In addition, another information search method of the present invention for solving the above-described problems includes a storage server that is connected to a network and stores data including document information, and data including the document information. An information search method used on a network system including a client including a browser for displaying a message and an analysis server including a morpheme analyzing unit, wherein the browser includes document information included in data displayed on the display unit Requesting the analysis server to perform a morphological analysis on the data, in response to a request from the analysis requesting means, the analysis server obtaining data displayed on the display means, and obtaining the data Cutting out words from the included document information by morphological analysis and related document information related to the words Searching from the storage server based on part-of-speech information and attribute information of a word; adding location information of the related document information to the acquired data; creating processed data; and And displaying data on the browser.

  According to the invention relating to the information retrieval system of the present application, both are connected to a network, a storage server in which data including document information is stored, a client including a browser that displays data including the document information, An information search system comprising an analysis server comprising morphological analysis means, wherein the browser performs execution of morphological analysis on document information included in data displayed on the display means. Analysis request means for requesting, and in response to a request from the analysis request means, the analysis server acquires data displayed on the display means, and from the document information included in the acquired data, a word is obtained by morphological analysis And searches the storage server for related document information related to the word, and the location of the related document information Information is added to the acquired data to create processed data, and the client displays the processed data on the browser. Therefore, processing such as adding related information to document information on the network in advance Therefore, it is possible to provide an information retrieval system that allows a user to immediately obtain related document information from a network at any time simply by selecting location information on a browser.

  Further, according to the invention relating to another information retrieval system of the present invention, a storage server that is connected to a network and stores data including document information, and a browser that displays the data including the document information An information retrieval system comprising: a client comprising: an analysis server comprising morphological analysis means; wherein the browser performs morphological analysis on document information contained in data displayed on the display means Analysis request means for requesting the analysis server, and in response to a request from the analysis request means, the analysis server acquires data displayed on the display means, and document information included in the acquired data A word is extracted from the word by morphological analysis, and related document information related to the word is based on the part of speech information and attribute information of the word. And searching the storage server, adding the location information of the related document information to the acquired data to create processed data, and the client displays the processed data on the browser. Information retrieval system that does not require processing such as adding related information to simple document information in advance, and the user can immediately obtain the optimum related document information from the network at any time simply by selecting the location information on the browser Can provide.

  According to the invention relating to the information search method of the present application, a client including a storage server that stores data including document information and is connected to a network, and a browser that displays the data including the document information. And an analysis server comprising a morphological analysis means, and an information search method used on a network system, wherein the browser performs morphological analysis on document information included in data displayed on the display means A step of requesting the analysis server; a step of acquiring the data displayed on the display unit in response to a request from the analysis request unit; and a morphological analysis from the document information included in the acquired data And extracting related words related to the word from the storage server. Adding the location information of the related document information to the acquired data to create processed data, and the client displaying the processed data on the browser. An information search method that does not require processing such as adding related information to the above document information in advance, and the user can immediately obtain related document information from the network at any time by selecting location information on the browser. Can be provided.

  Further, according to still another invention of the present application, both are connected to a network, a storage server in which data including document information is stored, a client including a browser for displaying data including the document information, An information search method used on a network system including a morpheme analysis unit, wherein the browser performs a morpheme analysis on document information included in data displayed on the display unit. A request to the analysis request unit, the analysis server acquiring data displayed on the display unit, and a word by morphological analysis from the document information included in the acquired data And the related document information related to the word as part of speech information and attributes of the word A step of searching from the storage server based on information, a step of creating processed data by adding location information of the related document information to the acquired data, and the client displaying the processed data on the browser The process of adding related information to a large amount of document information on the network in advance, and the user can select the optimum related document information simply by selecting the location information on the browser. An information search method that can be immediately obtained from the network at any time can be provided.

It is a figure which shows the basic composition of the information search system of this invention. It is a figure which shows a mode that the pseudo | simulation button as a morphological analysis request | requirement means is displayed on the browser screen of the pharmaceutical search system which is an Example of the information search system as one Embodiment of this invention with document information. It is a figure which shows a mode that processed data is displayed on the browser screen of the pharmaceutical search system which is an Example of the information search system as one Embodiment of this invention. It is a figure for demonstrating operation which acquires the related information by clicking the link of the processed data currently displayed on the browser screen of the pharmaceutical search system which is an Example of the information search system as one Embodiment of this invention. . It is a figure which shows the basic structure of the dictionary with which the morphological analysis means of the information search system of this invention is provided. It is a figure which shows an example of the more detailed structure of the dictionary with which the morphological analysis of the information search system of this invention is provided. It is a flowchart for demonstrating the main procedures of the information search method as one Embodiment of this invention.

  Embodiments of the present invention will be described with reference to the drawings. Hereinafter, an information search system according to one embodiment of the present invention will be described first, and then an information search method according to another embodiment of the present invention will be described with reference to a configuration example of the information search system. explain.

  FIG. 1 is a diagram showing a basic configuration of an information search system as an embodiment of the present invention. The information retrieval system 100 includes a client 120 including a browser 121, a storage server 130 storing data including document information, an analysis server including a morpheme analysis unit 141 and a dictionary 142, each connected to a network 110. 140 is configured as a computer network.

  The network 110 is a means for interconnecting the client 120, the analysis server 130, and the storage server 140 with a communication path. The communication path may be either wired or wireless, or may be configured using both. The scale of the network 110 is not limited, but is typically the Internet or an intranet in a company or the like. In the following description, it is assumed that the network 110 is exclusively the Internet.

  The client 120 is a means for requesting the storage server 130 and the analysis server 140 to perform a predetermined process and receiving the result of the process from the storage server 130 and the analysis server 140. The client 120 is a personal computer (personal computer) provided with the browser 121 or the like. Composed. In the actual information search system 100, although a plurality of clients 120 are often connected to the network 110 at the same time, only one client 120 is shown in FIG. 1 for convenience of explanation. .

  The browser 121 provided in the client 120 is a means for providing a function of displaying a web page including document information. Such browser functions are provided by, for example, “Internet Explorer (registered trademark)” of Microsoft Corporation. In addition, the browser 121 has a function of displaying a pseudo button 122 as an analysis request unit for starting the morphological analysis unit 141 of the analysis server 140 and causing the morphological analysis to be performed on the document information included in the web page. provide. The morphological analysis unit 141 and the pseudo button 122 as the analysis request unit will be described later in detail.

  The accumulation server 130 is means for accumulating data including document information, and is specifically a web server that accumulates web pages. When the web page, which is data including document information, is stored in the storage server 130 and made public, the web page is displayed on the browser 121 of the client 120 via the network 110 and browsed. In the actual information search system 100, a plurality of storage servers 130 are often connected to the network 110 at the same time. However, for convenience of explanation, only one storage server 130 is shown in FIG. Show only.

  In response to a request from the client 120, the analysis server 140 performs processing for searching related information related to the word included in the designated document information from the network 110, and also stores location information of the searched related information. Processing to add to the specified document information is performed, and the document information to which the location information of the related document is added is returned to the client 120 as a processing result. The analysis server 140 includes a morpheme analysis unit 141 as a unit for performing a process of extracting a word included in the document information. In addition, the morpheme analysis unit 141 includes a dictionary 142 used when performing morpheme analysis. Note that the analysis server 140 may include document search means 143 for executing a search process requested from the client 120.

  In response to a request from the client 120, the morpheme analysis unit 141 performs morpheme analysis on the document information included in the designated web page. A browser 121 of the client 120 displays a pseudo button 122 as an analysis request unit for causing a morphological analysis to be performed on document information included in the web page, along with the web page. When a user viewing a web page clicks the pseudo button 122, the web page displayed on the browser 121 is designated as a morphological analysis target, and the document information included in the web page is taken into the analysis server 140. The morpheme analysis unit 141 performs morpheme analysis.

  When the web page displayed on the browser 120 is designated by the pseudo button 122 as a morphological analysis target, the analysis server 140 acquires the content of the web page via the network 110. Here, when the analysis server 140 acquires a web page to be subjected to morphological analysis, only the location information such as the URL of the web page is acquired from the browser 121 and is used via the network 110 using the URL. It is efficient to configure to acquire web page data. Of course, the analysis server 140 can be configured to receive from the client 120 data of a web page to be subjected to morphological analysis.

  When the analysis server 140 acquires web page data to be subjected to morphological analysis, the analysis server 140 performs morphological analysis on the document information included in the data using the morphological analysis unit 141. And the analysis server 140 performs the process which adds the location information of the relevant information to each word cut out by the morphological analysis. Specifically, link destination information for linking each extracted word to related document information is searched, and processing for embedding the link destination information in the original web page data is performed. The web page data in which the link destination information is embedded (hereinafter referred to as processed data) is returned from the analysis server 140 to the client 120.

In the above processing, the morphological analysis unit 141 uses the dictionary 142 having part-of-speech information and attribute information for each word included. Therefore, the morpheme analyzing unit 141 can not only accurately extract each word from the document information based on the part of speech information and the attribute information, but can also search the network 110 for optimum related information for each extracted word. it can. When the field to which the document information to be processed by the analysis server 140 belongs is limited, the processing efficiency of the analysis server 140 can be further improved by using only words within a predetermined range of the dictionary 142. .

  The processed data returned from the analysis server 140 to the client 120 is displayed on the browser 121. At this time, the word in the web page displayed on the browser 121 displays that the above-mentioned link destination information is attached by an underline or the like. When the user of the information search system 100 selects a word to which the link destination information is attached by clicking on the browser 121, the related information existing at the link destination or the link destination information of the related information is selected. Is displayed on the browser 121. When link destination information of related information is displayed on the browser 121, the related information existing at the link destination is called and displayed on the browser 121 by selecting them by a method such as clicking them.

  Next, an information search method according to another embodiment of the present invention will be described with reference to the configuration of the information search system 100 described above. Hereinafter, the information search system 100 of the present invention will be described with reference to an embodiment in which the drug information search system is configured.

  FIG. 2 shows a state where an arbitrarily selected web page is displayed on the browser 121 as it is. In this state, the browser 121 displays a web page in which related information related to a certain medicine is described, and a pseudo button 122 for requesting execution of morphological analysis on document information in the web page. Yes. In this state, the user of the information search system 100 may move the pointer 123 and click the pseudo button 122 to learn more about the words included in the document information in the web page.

  In the browser 121, a pseudo button 122 serving as an analysis request unit for starting the morpheme analysis unit 141 of the analysis server 140 and executing morphological analysis on the document information included in the web page is configured as a part of the toolbar. Has been. This toolbar can be acquired from the analysis server 140 and may be acquired when the client 120 requests the analysis server 140 for the first time. The toolbar displays a pseudo button for displaying a predetermined home page, a pseudo button for automatically executing a morphological analysis when a web page is switched, and a word extracted by the morphological analysis means 141. A pseudo button or the like may be arranged together.

  FIG. 3 shows a display state of the web page after the pseudo button 122 displayed on the browser 121 is clicked. When the pseudo button 122 on the browser 121 is clicked, a word included in the document information in the web page is cut out as described above, and link destination information is added to each cut out word. Underlined. In this state, when the pointer 123 is moved over an underlined word (for example, “acetylsalicylic acid”), an index of related information related to the word is read from the dictionary 142 of the analysis server 140, for example, It is highlighted on the browser 121 by a pop-up 124 or the like.

  FIG. 4 shows a display state of the web page after the pop-up 124 displayed on the browser 121 is clicked. When the pop-up 124 on the browser 121 is clicked, the link destination information of the related information highlighted by the pop-up 124 is expanded, and detailed information is hierarchically displayed by the pop-up 125 or the like. In this state, when the pointer 123 is moved onto any link destination information (for example, “information providing site 3”) developed in the pop-up 125, related information existing in the link destination information is analyzed by the analysis server 140. Via the network 110 and displayed on the browser 121. Of course, when the link information of the related information is only one, the related information is acquired immediately by selecting the pop-up 124.

  Here, the dictionary 142 provided in the morphological analysis unit 141 of the analysis server 140 will be specifically described. The dictionary 142 is preferably a dictionary compiled for each field to which the information search system 100 is applied in order to increase the processing efficiency of the analysis server 140. FIG. 5 shows a part of the dictionary 142 configured in consideration of such points. Thus, the dictionary 142 may include not only the “medical field” but also a dictionary in a different field such as “architectural field” or “law field”. However, the dictionary that is most frequently used in the drug information search system is a dictionary in which the word “medical field” is stored, and the user selects the “medical field” dictionary from the client 120 in advance. Is desirable. In addition, when dictionaries of a plurality of fields are used, it is also effective to prioritize the dictionaries as necessary.

  The dictionaries of each field included in the dictionary 142 are hierarchized. For example, “medical field”, “architectural field”, etc. are arranged below “proprietary nouns”, and “ "1: Drug name" and "2: Injury name" are arranged. Further, the dictionary relating to “1: drug name” is further subdivided into “A: prescription drug”, “B: OTC (Over The Counter) drug”, “C: quasi drug”, “D: supplement”, etc. Has been. Similarly, the dictionary relating to “2: name of injury” is further subdivided into “A: internal medicine”, “B: surgery”, and the like. Furthermore, the dictionary of each field included in the dictionary 142 has part-of-speech information and attribute information for each stored word.

  FIG. 6 shows a partial internal structure of the “pharmaceutical name” dictionary shown in FIG. In the “pharmaceutical name” dictionary, for example, words such as “bufferin”, “bufferin A”, “bufferin plus”, “chocola BB”, “aminoprotein”, “revitalization” (all registered trademarks) are listed. ing. “Bufferin®”, “Bufferin A” and “Bufferin Plus” are trade names of Lion Corporation. “Chocola BB (registered trademark)” is a trade name of Eisai R & D Management Co., Ltd., and “Aminoprotein” and “Revitaly” are trade names. These words are common in that the part of speech is a proper noun. On the other hand, although the attributes of these words are common in that they are drug names, they are different in other respects. “Buffalin” is “A: prescription drug”, bufferin A ”and“ bufferin plus ”are“ “B: OTC pharmaceutical” and “Chocola BB” are “C: Quasi-drug”, “Aminoprotein” and “Revitaly” are “D: Supplement”.

  That is, FIG. 6 is a table in which “attributes” of words included in each dictionary are recorded together with “notation”, “reading”, “part of speech”, and the like. The morpheme analyzing means 141 of the analysis server 140 has the dictionary 142 hierarchized for each field, and the dictionary in each field has part-of-speech information and attribute information in the table format as described above. Such a data structure of the dictionary 142 enables the morphological analysis unit 141 to perform a process of extracting words (technical terms) in a specific field more accurately and efficiently.

  Next, the main procedure of the information search method of the present invention described above will be described with reference to the flowchart shown in FIG. First, it is assumed that a web page is displayed in the state shown in FIG. 2 on the browser 121 of the client 120 (step S0). Here, the web page displayed on the browser 121 may be acquired using a URL stored in the browser 121, or arbitrarily searched from the network 110 using the document search unit 143 of the analysis server 140. But you can.

  Next, when the pseudo button 122 displayed on the browser 121 is clicked, execution of morphological analysis is instructed from the client 120 to the analysis server 140 (step S1). Based on this instruction, the web page displayed on the browser 121 is designated as a morphological analysis target, and the analysis server 140 takes in the content of the web page via the network 110 (step S2).

  Next, the morphological analysis unit 141 of the analysis server 140 performs morphological analysis on the document information included in the fetched web page, and cuts out words (step S3). Then, the analysis server 140 searches for link destination information for linking the extracted words to related document information, and embeds the link destination information in the acquired web page data (S4). This embedded data is used when displaying related information on the browser 121 later.

  The processed data of the web page in which link destination information is added to each word included in the document information by the analysis server 140 is returned to the client 120 and displayed as a candidate on the browser 121 (step S5). When the link destination information added to each word included in the processed data is selected by the user (step S6), a process for causing the browser 121 to display related information associated with the link destination information is instructed. The link destination information is sent to the storage server 130 to identify the related information to be displayed (step S7). Then, the identified related information is sent from the storage server 130 to the browser 121 and displayed on the browser 121 (step S8).

  If the user determines that the target search result is obtained, the search is terminated. Otherwise, the steps from step S1 to step S8 are repeated until the target search result is obtained.

  As described above, according to the present invention, the analysis server extracts words from document information included in the data displayed on the browser by morphological analysis, and retrieves related document information related to the extracted words. It is not necessary to perform processing such as adding related information to the document information on the network in advance, and the user can select the location information on the browser. An information search system and an information search method that can be immediately acquired from a network can be provided.

  In the above description, the drug information search system has been referred to. However, the application of the information search system and information search method of the present invention is not limited to this, and as an information search system and information search method in all fields. It is feasible. In addition, the information search system and information search method of the present invention can be implemented not only on a large-scale network such as the Internet but also on a small and medium-sized network such as an intranet or a local area network.

100 Information Retrieval System 110 Network 120 Client 121 Browser 122 Pseudo Button 130 Storage Server 140 Search Server 141 Morphological Analysis Means 142 Dictionary

Claims (4)

Each includes a storage server that is connected to a network and stores data including document information, a client that includes a browser that displays data including the document information, and an analysis server that includes morphological analysis means. An information retrieval system comprising:
The browser comprises analysis request means for requesting the analysis server to perform morphological analysis on document information included in data displayed on the display means;
In response to a request from the analysis request unit, the analysis server
Obtaining data displayed on the display means;
Extract words by morphological analysis from the document information included in the acquired data,
Search related document information related to the word from the storage server,
Adding location information of the related document information to the acquired data to create processed data;
The information search system, wherein the client displays the processed data on the browser.   The information search system according to claim 1, wherein the analysis server searches for the related document information based on part-of-speech information and attribute information of the word. A network including a storage server that stores data including document information, a client that includes a browser that displays data including the document information, and an analysis server that includes morphological analysis means, all connected to the network. An information search method used on a system,
Requesting the analysis server to perform morphological analysis on document information included in data displayed on the display means by the browser;
In response to a request from the analysis request unit, the analysis server
Obtaining data displayed on the display means;
Extracting words from document information included in the acquired data by morphological analysis;
Searching the storage server for related document information related to the word;
Adding location information of the related document information to the acquired data to create processed data;
The client displaying the processed data on the browser;
A method for retrieving information, comprising:   4. The information search method according to claim 3, wherein the related document information is searched based on part-of-speech information and attribute information of the word.