JP2007011973A - Information retrieval device and information retrieval program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information retrieval device capable of increasing an information amount related to a retrieval character string without increasing an operation amount. <P>SOLUTION: This information retrieval device has: an input device 2 functioning as a retrieval character string input part for inputting the retrieval character string; an information analysis device 1 functioning as a retrieval target information acquisition part acquiring retrieval target information, a retrieval character string information extraction part extracting information related to the retrieval character string from the retrieval target information, a related term extraction part extracting a related term having prescribed correspondence relation with the retrieval character string from the retrieval target information on the basis of the information extracted by the retrieval character string information extraction part, and a retrieval result generation part generating a retrieval result by use of the related term extracted by the related term extraction part; and a display device 3 displaying the retrieval result. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an information search apparatus for searching for search target information and an information search program for causing a computer to search for search target information.

  With the spread of the Internet, the scale of search target information handled by general clients has increased dramatically. As a result, the importance of a method for selecting information required by the person is increasing.

  A search system called a search engine is most used as an information search method on the Internet. One of the most powerful search engines is the full-text search type. A full-text search type search engine is an information search system that saves the contents of a Web page in a database on the search engine side, searches the database when a search request is made, and pops out the result. Most of the results output by a full-text search type search engine are simply extracted search strings.

  Even if a conventional general information search system adds some processing to the search result, it is often necessary to obtain more information from the client, and the processing becomes more or less complicated. Furthermore, from the viewpoint of searching for a wide range of knowledge, we had to rely on the primitive method of inputting a new search string by the client.

  In addition to the conventional general search system described above, various information search systems have been proposed.

  An example of an information search system that can improve the accuracy and reliability of a search result is disclosed in Patent Document 1. In the search system disclosed in Patent Document 1, the accuracy and reliability of the search results are improved by providing a dictionary database that records words representing matters that the questioner wants to know.

  An example of an information search system that can extract a portion related to a search character string from a search document is disclosed in Patent Document 2. The information retrieval system disclosed in Patent Document 2 accumulates the degree of relevance of related keywords for each sentence of a conforming document, and determines the order of appearance in the document until a predetermined length is reached in order from the sentence with the largest accumulated value. Extract sentences while keeping them.

An example of an information search system that can display search information with superiority or inferiority is disclosed in Patent Document 3. The information retrieval system disclosed in Patent Document 3 performs syntax analysis on the retrieved text, extracts keywords, examines the syntax information of the keywords, and determines the conditions under which the keyword syntax information is set. The priority order is determined depending on whether or not it is satisfied, and the priority order is given to the retrieved text.
JP 2002-123541 A JP 2001-84255 A Japanese Patent Laid-Open No. 6-215035

  In the information search system disclosed in Patent Document 1, there is a difference in the accuracy and reliability of the search results depending on the scale of the dictionary database that contains words representing the matters that the questioner wants to know.

  In addition, although the information search system disclosed in Patent Document 2 extracts sentences from the viewpoint of the degree of relevance, it extracts sentences while maintaining the order in which they appear in the document in descending order of the relevance accumulated value. Therefore, the reusability as data in the analysis result is poor, and it is difficult to draw the information of the important search character string itself.

  The information search system disclosed in Patent Document 3 determines the priority order based on whether or not the keyword syntax information satisfies a set condition, and gives the priority order to the searched text. Depending on the set conditions, there are some texts that receive an unreasonably low rating, making it difficult to achieve a consistent evaluation standard.

  As described above, the information search systems disclosed in Patent Documents 1 to 3 try to provide more dense information by reducing a considerable amount of data in the process of processing data, but are applied in that process. Depending on the criteria used, an inefficient search may occur.

  Also, in a conventional general information search system, when trying to search a wide range of knowledge about search terms according to the input by the client, the search result is obtained once, and the related terms are extracted from it. It was necessary to take a procedure such as searching again. Regarding search actions, it is important to acquire as much necessary knowledge as possible with few steps and simple operations, so it is important to request new input from the client and to set up complicated search procedures. It should be avoided to set criteria that may contain search results or reduce search results.

  In view of the above problems, an object of the present invention is to provide an information search apparatus and an information search program that can increase the amount of information related to a search character string without increasing the amount of operation.

  In order to achieve the above object, an information search apparatus according to the present invention includes a search character string acquisition unit that acquires a search character string, a search target information acquisition unit that acquires search target information, and the search target information. A search character string information extracting means for extracting information related to the search character string, and a predetermined correspondence between the search character string and the search character string based on the information extracted by the search character string information extracting means. The system includes a related phrase extracting unit that extracts a related phrase having a relationship, and a search result generating unit that generates a search result using the related phrase extracted by the related phrase extracting unit.

  According to such a configuration, as in the conventional general full-text search type information search system, the information search is executed only by the input operation of the search character string, and there is a predetermined correspondence with the search character string. Therefore, it is possible to increase the amount of information related to the search character string without increasing the amount of operation.

  Further, the information processing apparatus further includes additional information generation means for generating additional information included in the search result, and the additional information generation means includes a summary sentence including the related phrase based on the related phrase extracted by the related phrase extraction means. You may make it be one element of the said additional information. Further, the additional information generating means may include a summary sentence including the related word phrase and the search character string based on the related word phrase extracted by the related word phrase extraction means and the information extracted by the search character string information extraction means. You may make it be one element of the said additional information.

  According to such a configuration, since the amount of information in the search result alone increases, the client can acquire a lot of necessary knowledge.

  Further, it comprises additional information generating means for generating additional information included in the search result, and the additional information generating means is extracted by the related phrase extracting means based on the related phrases extracted by the related phrase extracting means. A storage location of information including related phrases may be an element of the additional information.

  According to such a configuration, the client can move to a place with necessary information with a simple and few operations, and the efficiency of the search process can be improved.

  Further, the information processing apparatus includes additional information generation means for generating additional information included in the search result, wherein the additional information generation means has a relevance degree that is information obtained by quantifying the relation between the related phrase and the search character string. You may make it be one element of the said additional information.

  According to such a configuration, since the amount of information in the search result alone increases, the client can acquire a lot of necessary knowledge.

  Moreover, you may provide the related phrase sorting means which rearranges the said related phrases on the basis of the said related degree. Further, the additional information generating means may calculate the degree of association based on the number of characters existing between the related phrase and the search character string, and the same phrase of the related phrase extracted by the related phrase extracting means The degree of association may be calculated based on the total number of.

  In order to achieve the above object, an information search program according to the present invention provides a computer with search character string acquisition means for acquiring a search character string, search target information acquisition means for acquiring search target information, Search character string information extracting means for extracting information related to the search character string from the inside, based on the information extracted by the search character string information extracting means, a predetermined space between the search character information and the search character string Related phrase extracting means for extracting a related phrase having a corresponding relationship, and an information search program for functioning as a search result generating means for generating a search result using the related phrase extracted by the related phrase extracting means It is said.

  According to the present invention, it is possible to realize an information search device and an information search program that can increase the amount of information related to a search character string without increasing the amount of operation.

  Embodiments of the present invention will be described below with reference to the drawings. In the following description, the same parts and components are denoted by the same reference numerals, and detailed description of the same parts and components will not be repeated.

<Definition of terms>
First, definitions of terms used in the description of the embodiments of the present invention will be described.

  “Client” refers to a person who is in a position to request a search process to the information search apparatus according to the present invention. For example, a person who actually uses the information search apparatus according to the present invention transmits a request for the search process. This corresponds to an internal / external arithmetic processing device, an arithmetic processing program, and the like.

  `` Exists in the vicinity '' means that multiple characters are located close to each other (for example, in the same sentence), or multiple characters are conceptually close to each other. A case where there is some kind of correspondence between columns.

  A “related phrase” refers to a character string having some relationship with a certain character string. There is no problem that the related phrase includes the search character string itself. For example, there is no problem in using the word “guide dog” extracted by the search character string “dog” as the related word phrase of the search character string “dog”.

  “Recording means” refers to means for recording information. For example, a main storage device represented by flash memory, an external storage device represented by HDD (Hard Disk Drive), a CD-ROM (Compact Disk Read Only Memory). And recording media such as memory cards.

  The “input device” refers to a device that gives information corresponding to a human input operation to an arithmetic device such as a CPU, and may be in any form such as a keyboard, a mouse, and a tablet.

  “Input information” refers to at least one of information input by a client and information stored in a recording unit. The “information input by the client” corresponds to, for example, character information input from a keyboard, stroke information input from a tablet, character information obtained by character recognition processing of the stroke information, and the like. In addition, an image obtained by photographing a character written on paper with a camera or the like, or character information obtained by character recognition of the image can be considered.

  “Notification means” refers to means for notifying information provided from an arithmetic device such as a CPU. Not only a display device that displays information provided from an arithmetic device such as a CPU, but also a signal such as a vending machine. A device that receives the goods and receives the goods also corresponds to the notification means.

  “Communication device” refers to a device installed between an arithmetic device such as a CPU and the outside for the purpose of receiving information from outside that is not inside the information search device. The form of the gateway device, router, dedicated line, etc. It doesn't matter.

  “Search target information” mainly refers to electronically recorded data or the like, and may be any data that can be expressed in a form that can be normally recognized by a person. The format that can be recognized by humans means, for example, a format in which characters and images are displayed on a display or printed on paper. The contents of the data include all kinds of data that can be expressed in a format that can be recognized by the human senses using the five senses, such as touch, smell, and taste, as well as visual and auditory senses such as characters, figures, images, videos, and sounds. Examples of search target information include business document files used in offices, electronic books, and electronic data on the Internet. Of course, data that cannot be handled as a normal character string, such as music and movies, may be searched using a special technique, and is included. Usually, the search target information is often recorded in a recording means in the form of a file or the like. Various formats can be considered, but for example, file formats and database formats are common. The reading / writing of the search target information may be directly performed on a device in the information search apparatus, or may be indirectly performed on a device having a different layer through a communication line.

  The “index” refers to an index indicating position information such as which line is in order to improve the information retrieval speed. The format of the index does not need to be in units of lines, and any index can be used as long as it can express the location of data such as byte positions and URLs.

  “Morphological analysis” refers to the process of dividing a given sentence into morphemes, and “morpheme” refers to the smallest unit that does not change in a word. The morpheme analysis also includes correspondence as to which part of speech each morpheme corresponds to. In the present invention, in the morpheme analysis, the part that decomposes the sentence into morphemes and the part that associates the part of speech with each other are taken up. Paradoxically speaking, the morpheme segmentation is performed without using a strict morphological analysis method. In the present invention, there is no problem even if the discrimination method is adopted instead of the morphological analysis method.

<Outline of Information Retrieval Device According to the Present Invention>
Next, an outline of the information search apparatus according to the present invention will be described. An example of the appearance of the information retrieval apparatus according to the present invention is shown in FIG. The information search device shown in FIG. 1 includes an information analysis device 1, an input device 2 that gives information necessary for the search to the information analysis device 1, and a display device 3 that displays the search results received from the information analysis device 1. .

  1, the information analysis device 1, the input device 2, and the display device 3 are separate from each other. However, the information search device according to the present invention is not limited to this, and the information search device The analysis device 1, the input device 2, and the display device 3 may be integrally configured. Further, other informing means may be used instead of the display device 3. For example, when the search target information is music data, it is conceivable to use an audio output device instead of the display device 3.

  The information search apparatus according to the present invention does not output the search character string as it is in response to the input of the search character string by the client. The search string on the search target information specified by the full-text search method does not exist by itself, but most of the search string exists in the sentence and a number of phrases are placed around it. is doing. It is considered that there is a strong semantic connection between the surrounding words and the search character string, including a context relationship.

  Therefore, in the information search apparatus according to the present invention, the phrases existing around these are collected as the related phrases of the search character string and output as search results. An actual display example is shown in FIG. In the display example illustrated in FIG. 2, “cat”, “pet”, “service dog”, and “guide dog” that are related words in the search character string “dog” are listed.

  Search results are compiled based on the same related terms and are output for each group. The search result mentioned here is a summary of the sentence including the search string, the location where the related words are stored, the frequency of occurrence of related words quantified based on the collected data, etc., or a combination of these, There is no particular limitation as long as it can be created using the obtained data. An actual display example is shown in FIG. In the display example shown in FIG. 3, “cat” and “pet”, which are related terms of the search character string “dog”, and a summary sentence that is the additional information, and a relationship between the search character string and the related terms are quantified , And Internet addresses are displayed.

<Details of Information Retrieval Device According to the Present Invention>
Next, details of the information retrieval apparatus according to the present invention will be described.

  The function of the information search apparatus shown in FIG. 1 will be described with reference to FIG. FIG. 4 is a functional block diagram of the information search apparatus shown in FIG. 1 includes a search character string input unit 4, a search target information acquisition unit 5, an index generation unit 6, a search character string information extraction unit 7, a related word / phrase extraction unit 8, and a related word / phrase sort unit. 9, an additional information generation unit 10, a search result generation unit 11, and a search result output unit 12. Search target information acquisition unit 5, index generation unit 6, search character string information extraction unit 7, related word / phrase extraction unit 8, related word / phrase sort unit 9, additional information generation unit 10, and search result generation unit 11 Is realized by the information analysis device 1, the search character string input unit 4 is realized by the input device 2, and the search result output unit 12 is realized by the display device 3.

  The search character string input unit 4 sends information given from the client to the search character string information extraction unit 7.

  The search target information acquisition unit 5 acquires search target information stored in an internal or external recording unit of the information search apparatus, and sends the acquired search target information and the like to the search character string information extraction unit 7. Further, in order to create an index when the search target information is enormous, the search target information acquisition unit 5 sends the acquired search target information to the search character string information extraction unit 7 via the index generation unit 6. send.

  The search character string information extraction unit 7 receives information sent from the search character string input unit 4 and the search target information acquisition unit 5 or the index generation unit 6, and the position (address, etc.) of data including the search character string And extract indexes.

  The related word / phrase extraction unit 8 extracts words / phrases existing around the search character string as related words / phrases based on the position (address, etc.) and index of the data including the search character string extracted by the search character string information extraction unit 7. To do.

  The related phrase sorting unit 9 sorts the related phrases extracted by the related phrase extracting unit 8 according to a rule such as dictionary order. The rules used here may be in any form as long as the client can easily use the search results.

  The additional information generating unit 10 generates additional information that can be generated from the information extracted by the search character string information extracting unit 7 and the related phrase extracting unit 8.

  The search result generation unit 11 adds the additional information generated by the additional information generation unit 10 to the information rearranged by the related phrase sorting unit 9, and generates a search result for notification to the client.

  The search result output unit 12 displays the search result generated by the search result generation unit 11 and notifies the client of information obtained by the search.

  In the above description, there is a part that expresses data etc. in the verb “send” in the sentence, but this does not mean only when the subject actually works on the object, but the object Even when the part that works is the subject, and the data is obtained from the subject part, it is expressed by the word “send” for convenience.

  Next, the configuration of the information search apparatus shown in FIG. 1 will be described with reference to FIG. FIG. 5 is a diagram illustrating a configuration example of the information search apparatus illustrated in FIG. The information analysis device 1 includes a main storage device 1A, an external storage device 1B, a CPU (central processing unit) 1C, a communication device 1D, and a data bus 1E. The data bus 1E allows the input device 2 and display. Data exchange with the device 3 is performed.

  The main storage device 1A is a temporary storage device such as a flash memory, and includes a search target information acquisition unit 5, an index generation unit 6, a search character string information extraction unit 7, a related phrase extraction unit 8, a related phrase sorting unit 9, and additional information. The main purpose is to store data acquired, extracted, or generated by the generation unit 10 and the search result generation unit 11. The register area included in the CPU (Central Processing Unit) 1C is also shared with the main storage device 1A, and the search target information acquisition unit 5, the index generation unit 6, the search character string information extraction unit 7, and the related phrase extraction unit. 8. The data acquired, extracted, or generated by the related phrase sorting unit 9, the additional information generation unit 10, and the search result generation unit 11 is stored.

  The external storage device 1B is a removable storage device such as an HDD (Hard Disk Drive), and an analysis result calculated by a CPU (Central Processing Unit) 1C or a save target that is temporarily unnecessary for the main storage device 1A. The main purpose is to store the data. It is also possible to use a main storage device or an external storage device outside the information search device that can be used by wired connection or wireless connection via the communication device 1D and the network 13 as the external storage device 1B.

  The CPU (central processing unit) 1C includes a search target information acquisition unit 5, an index generation unit 6, a search character string information extraction unit 7, a related phrase extraction unit 8, a related phrase sorting unit 9, an additional information generation unit 10, and a search result. A program describing a procedure of data processing performed by the generation unit 11 is acquired from the main storage device 1A, the external storage device 1B, the network 13 via the communication device 1D, or the like, and the search target information acquisition unit 5, the index generation unit 6, Data processing performed by the search character string information extraction unit 7, the related phrase extraction unit 8, the related phrase sorting unit 9, the additional information generation unit 10, and the search result generation unit 11 is executed. Note that data exchange between the CPU (central processing unit) 1C and other devices is not limited to the data bus 1E, and any data transmission / reception device such as a wired or wireless communication device can be used. Good. Further, data processing performed by the search target information acquisition unit 5, the index generation unit 6, the search character string information extraction unit 7, the related phrase extraction unit 8, the related phrase sorting unit 9, the additional information generation unit 10, and the search result generation unit 11 Is not limited to the CPU (central processing unit) 1C, and any device may be used as long as the data processing can be executed.

  The display device 3 functions as the search result output unit 12, displays the search result generated by the search result generation unit 11, and notifies the client of information obtained by the search.

  The input device 2 functions as the search character string input unit 4 and supplies information (search character string) given by the client to the CPU (central processing unit) 1C. The format of the input device 2 is not particularly limited, and any format can be used as long as a search character string can be described.

  The communication device 1C exchanges data with other network devices by wired connection or wireless connection.

  Next, a data processing procedure in the information search apparatus having the configuration shown in FIG. 5 will be described with reference to FIG. FIG. 6 is a flowchart showing an example of a data processing procedure in the information search apparatus having the configuration shown in FIG.

  The data processing shown in FIG. 6 is realized by the CPU (central processing unit) 1C reading and executing a program stored in the main storage device 1A, the external storage device 1B, etc., and controlling the entire information retrieval device.

  In step S10, the input device 2 functioning as the search character string input unit 4 sends a character string given from the client by the user's input operation as a client to the CPU (central processing unit) 1C as a search character string.

  In step S <b> 20, the CPU (central processing unit) 1 </ b> C functions as the search target information acquisition unit 5 and the search character string information extraction unit 7. Specifically, the CPU (central processing unit) 1C acquires information stored in the external storage device 1B or the like, or search target information obtained from the network 13 via the communication device 1D, and stores the search target information. The data position of the search character string is extracted from the inside, and the data position of the extracted search character string is written into its own register area or the main storage device 1A or the external storage device 1B. At this time, the data position of the extracted search character string is stored in the array index [k] (k is an arbitrary natural number of 1 ≦ k ≦ n) shown in FIG.

  An example of the internal structure of data obtained by the data processing procedure shown in FIG. 6 is shown in FIG. FIG. 8A shows an array index [k] that holds the data position of the search character string. FIG. 8B shows a word (element) extracted as a related phrase and a word that is found as a result of analyzing the sentence. This is a structure relation that secures the part of speech (attribute) and the position (index) of the word. FIG. 8 (c) shows the combination of the word (element) extracted as the related phrase and its additional information (add_info). Structure addition to be held.

  FIG. 9 is a schematic diagram showing the internal structure of the data shown in FIG. 8 with a focus on the hierarchical structure. Each index [k] holds a set of structure relations as its child elements. Each of these structure relations includes a word (element) extracted as a related phrase, and one-to-one or one-to-many additional information (add_info) is extracted from the word (element) extracted as the related phrase. The child element has a structure addition that contains. The data structure of the hierarchical structure shown in FIG. 9 is considered to be an effective aspect with respect to the present embodiment, and it is possible to realize the information retrieval apparatus according to the present invention using other data structures. is there. For example, even in a mode in which data is sequentially stored in the order in which data is acquired in a single record as shown in FIG. 10A, a record in which history is recorded as shown in FIG. 10B is prepared. May be.

  In step S30, the CPU (central processing unit) 1C functions as the related phrase extraction unit 8. That is, the CPU (central processing unit) 1C extracts related words based on the data position of the search character string extracted in step S20. Details of the processing in step S30 will be described with reference to FIG. FIG. 7 is a flowchart illustrating an example of a related word / phrase extraction processing procedure.

  In step S31, the CPU (central processing unit) 1C sets the array index [k] that holds the data position of the search character string shown in FIG. 8B to its own register area or the main storage device 1A or the external storage device 1B. Read from.

  In step S32, the CPU (central processing unit) 1C specifies the work area of the search target information using the data position of the search character string acquired in step S31.

  In step S33, the CPU (central processing unit) 1C extracts a sentence including the search character string from the work area of the search target information specified in step S32. Here, the “sentence” refers to a portion delimited by the punctuation mark “.”.

  In step S34, the CPU (central processing unit) 1C decomposes the extracted sentence by a morphological analysis method, stores the segmented word in the element of the structure relation shown in FIG. 8B, and stores the part of speech of the segmented word. It is stored in the attribute of the structure relation shown in FIG. 8B, and the data position of the segmented word is stored in the index of the structure relation shown in FIG. 8B. Normally, ambiguity is the subject of debate in morphological analysis methods, but there are few sentence examples where multiple candidates are calculated when dealing with mixed sentences of Kana-Kanji characters, and in English, decomposition is done in units of words. Therefore, it may be considered that the processing of step S34 is not hindered.

  In step S35, the CPU (central processing unit) 1C has the word of the structure relation only when a specific value (eg, noun) is included in the part of speech (attribute) of the structure relation shown in FIG. (Element) is a related phrase, and a structure relation that does not satisfy this condition is deleted.

  In step S36, the CPU (central processing unit) 1C confirms whether there are any elements in the array index [k] shown in FIG. 8A that have not been subjected to the processes in steps S32 to S35. If there is a target element (an element that has not been subjected to the processes in steps S32 to S35) (NO in step S36), the process returns to step S32. If there is no target element (YES in step S36), the process exits this flowchart and proceeds to step S40 shown in FIG.

  Referring to FIG. 6 again, the description of the data processing procedure in the information search apparatus having the configuration shown in FIG. 5 will be continued.

  In step S40, the CPU (central processing unit) 1C converts a sentence including the word (element) of the structure relation illustrated in FIG. 8B, that is, a sentence including a related phrase into the structure relation illustrated in FIG. 8B. Is extracted from the search target information using the array (index) and stored in add_Info of the structure addition shown in FIG. In the element of the structure addition shown in FIG. 8C, the related phrase that is currently used as the extraction criterion is stored as it is.

  In step S50, the CPU (central processing unit) 1C confirms whether there is an element from which additional information is not extracted in the array index [k] illustrated in FIG. If there is a target element (an element not subjected to the process of step S40) (NO in step S50), the process returns to step S40. If there is no target element (YES in step S50), the process proceeds to step S60.

  In step S60, the CPU (central processing unit) 1C calculates a specific reference for the element and add_Info of the structure addition shown in FIG. 8C (for example, in the order of the dictionary of elements), and based on the calculation result, The order of the structure relations shown in FIG. 8B is changed and stored again in the main storage device 1A, the external storage device 1B, or the like.

  In step S70, the CPU (central processing unit) 1C generates data as a search result based on the structure addition shown in FIG. 8C, sends the search result to the display device 2, and the flow shown in FIG. Ends.

  The following are specific examples when the search results are actually output and displayed. FIG. 2 shows only the related terms arranged as an output result when “dog” is designated as a search character string. The order in which the related phrases are arranged includes, for example, the order found from the search target information, the dictionary order, the order in which the number of searched items is large, and the like.

  FIG. 11 shows a summary sentence as additional information in addition to a related phrase when “dog” is designated as a search character string. As a result, the amount of information in the search result alone increases, and the client can easily obtain necessary information from this screen. As a summary sentence extraction method, for example, a sentence that includes a search character string, a sentence that includes a related phrase, or a sentence that includes both of them is extracted as a summary sentence. In addition, for the setting related to the length of the summary sentence to be extracted, for example, a sentence with a punctuation mark (a symbol such as a period in English) can be used as a summary sentence, or a threshold value can be set within the range. Settings that extract as much as possible are listed. Further, as shown in FIG. 11, when a summary text is displayed, if a search character string included in the summary text or a related word / phrase is highlighted in bold or bold, it is considered that the display is easy to see for the client.

  FIG. 12 shows, as additional information, an Internet address at which relevant data is posted in addition to related terms when “dog” is designated as a search character string. URL (Uniform Resource Locator) is listed as an output example, but data is simply stored by listing links specified for different character strings using HTML (Hyper Text Markup Language) description. Any method can be used as long as it can point to the place where the information is being placed. As a result, the client can move to a place with necessary information with a simple and few operations, and the efficiency of the search process can be improved.

  FIG. 13 shows a case where “dog” is designated as a search character string, and in addition to related terms, the degree of association between the two is displayed as additional information. As a result, the amount of information in the search result alone increases, and the client can easily obtain necessary information from this screen. In this output example, the number of related terms actually searched is used as a method of calculating the degree of association, and the numbers are displayed in descending order. It is not necessary to display all the related phrases searched, and only a predetermined number (for example, three) may be displayed in order from the largest number of cases. That is, there is no need to limit the number of related phrases displayed.

  FIG. 14 shows a case where “dog” is designated as a search character string, and the proximity of both of them is displayed as additional information in addition to related terms. As a result, the amount of information in the search result alone increases, and the client can easily obtain necessary information from this screen. In this output example, the proximity is determined by the number of characters existing between the search character string and the related phrase corresponding thereto. In addition, the percentage is displayed horizontally as a reference for arranging related terms, but for this calculation method, among the extracted related terms with the same name, the ratio of the set whose proximity is less than a certain value, Instead of displaying the percentage itself, it is possible to adopt a method of simply taking the average of each proximity. Further, as an average calculation method, for example, there is a method of taking an average of all searched items, a method of extracting a predetermined number (for example, 10) in ascending order of proximity, and taking an average thereof. .

These are figures which show the example of an external appearance of the information search device which concerns on this invention. These are figures which show the example of an output display of the information search device which concerns on this invention. These are figures which show the example of an output display of the information search device which concerns on this invention. These are functional block diagrams of the information search apparatus shown in FIG. These are figures which show the example of 1 structure of the information search device shown in FIG. These are the flowcharts which show an example of the data processing procedure in the information search device of a structure shown in FIG. These are the flowcharts which show an example of the extraction processing procedure of a related term phrase. These are figures which show the internal structure of the data obtained by the data processing procedure shown in FIG. FIG. 7 is a schematic diagram showing an internal structure of data obtained by the data processing procedure shown in FIG. 6. These are figures which show the other structure of the data obtained by the data processing procedure shown in FIG. These are figures which show the example of an output display of the information search device which concerns on this invention. These are figures which show the example of an output display of the information search device which concerns on this invention. These are figures which show the example of an output display of the information search device which concerns on this invention. These are figures which show the example of an output display of the information search device which concerns on this invention.

Explanation of symbols

1 Information Analysis Device 1A Main Storage Device 1B External Storage Device 1C CPU (Central Processing Unit)
1D Communication Device 1E Data Bus 2 Input Device 3 Display Device 4 Search Character String Input Unit 5 Search Target Information Acquisition Unit 6 Index Generation Unit 7 Search Character String Information Extraction Unit 8 Related Phrase Extraction Unit 9 Related Phrase Sorting Unit 10 Additional Information Generation Unit 11 Search Result Generation Unit 12 Search Result Output Unit 13 Network

Claims (9)

A search string acquisition means for acquiring a search string;
Search target information acquisition means for acquiring search target information;
Search character string information extracting means for extracting information related to the search character string from the search target information;
Based on the information extracted by the search character string information extraction unit, a related word phrase extracting unit that extracts a related word phrase having a predetermined correspondence with the search character string from the search target information; ,
An information search apparatus comprising: a search result generation unit that generates a search result using the related phrase extracted by the related phrase extraction unit. Comprising additional information generating means for generating additional information included in the search results;
The information search device according to claim 1, wherein the additional information generation unit uses a summary sentence including the related phrase as one element of the additional information based on the related phrase extracted by the related phrase extraction unit.   The additional information generation unit adds the summary sentence including the related phrase and the search character string based on the related phrase extracted by the related phrase extraction unit and the information extracted by the search character string information extraction unit. The information search apparatus according to claim 2, wherein the information search apparatus is an element of information. Comprising additional information generating means for generating additional information included in the search results;
The storage location of information including the related phrase extracted by the related phrase extracting unit based on the related phrase extracted by the related phrase extracting unit as one element of the additional information. The information search apparatus in any one of 1-3. Comprising additional information generating means for generating additional information included in the search results;
The information according to any one of claims 1 to 4, wherein the additional information generation unit uses, as one element of the additional information, a degree of relevance that is information obtained by quantifying the relevance between the related phrase and the search character string. Search device.   The information search device according to claim 5, further comprising: a related phrase sorting unit that rearranges the related phrases based on the relevance level.   The information search device according to claim 5 or 6, wherein the additional information generation unit calculates the degree of association based on the number of characters existing between the related phrase and the search character string.   The information search apparatus according to claim 5 or 6, wherein the additional information generation unit calculates a relevance level based on a total number of identical phrases of the related phrases extracted by the related phrase extraction unit. Computer
Search string acquisition means for acquiring a search string,
Search target information acquisition means for acquiring search target information,
Search character string information extracting means for extracting information related to the search character string from the search target information;
Based on the information extracted by the search character string information extraction unit, a related word phrase extraction unit that extracts a related word phrase having a predetermined correspondence with the search character string from the search target information; And an information search program for functioning as search result generation means for generating a search result using the related phrases extracted by the related phrase extraction means.

Images