JP2007279978A - Document retrieval device and document retrieval method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To retrieve documents having similar formats by checking the matching of the format information of the documents and evaluating similarity. <P>SOLUTION: The document retrieval device comprises an input/output device, a document database 105 for storing the documents, a document format information extraction part 107 for extracting the format information of the documents, a similarly configured document retrieval part 108 for retrieving the documents similar to the format of the document desired to be retrieved or a retrieval condition specified by a user, and a retrieved result display part 109 for outputting retrieved results to the input/output device. A means for calculating a similarity degree indicating the degree of the similarity of the document on the basis of a weight value attached to each format information is provided and the retrieved result is displayed by the level of the similarity degree. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a document search apparatus and a document search method for searching for documents using a computer.

  As a document search method using a computer, based on the documents or keywords entered as search conditions, the similarity of the search target document group is determined using syntax analysis, phrase frequency, etc. There is a way to search. In the conventional search method, a weight calculated based on the appearance frequency or the like is assigned to a keyword extracted from a search condition or a search target document, the similarity is quantified from the weight, and a document having a high similarity is output as a search result. .

  Patent Document 1 discloses an example of a similar document search method that performs keyword extraction and keyword weighting by paying attention to a scale expression word.

JP 2005-301855 A

  At the time of document retrieval, the semantic information of the document is an important element, but there is a case where the style of the document created or browsed in the past is stored in the memory, and it is desired to search based on it. For example, when a new document is created using a report format created by another person in the past as a model, it is desired to perform a search according to the document style without depending on the contents of the past report. However, in the conventional search method, the content of the document is interpreted and the similarity is evaluated semantically. Therefore, it is not possible to search for a similar document based on the format information of the document. On the other hand, for CAD data and the like, there is a search method for recognizing a search target as “shape” and calculating similarity based on geometric shape feature data, but it is not applicable to document search.

  The present invention has been made in view of such a point, and format information of a document, that is, a template used by the document, a set property, a document type, a file name, a header and a footer, a frame, and the like. An object of the present invention is to search for documents having similar formats by inspecting matching and mismatching of ruled line patterns, paper sizes, columnar forms, layouts, page number information, and the like to evaluate similarities.

  The present invention relates to a document search apparatus using a computer, an input / output device such as a display, a keyboard, and a mouse, a database that stores documents, a document format information extraction unit that extracts document format information, A similar-structured document search unit that searches the database for a document to be searched or a document similar to the search condition format specified by the user, and a search result display unit that outputs the search result to the input / output device are provided. Furthermore, a means for weighting each format information and a means for calculating a similarity indicating the degree of similarity between documents are provided, and the search results are displayed in order of the degree of similarity.

  With this configuration, when searching for a document similar to the input document, the format information of the input document, that is, the template used by the document, the set property, and the document type , File name, header and footer, frame and ruled line pattern, paper size, columnar form, layout, page number information, etc. are extracted from the input document, and the input document and format from the document database It is possible to search for documents that are similar to each other, that is, documents in which some format information among a plurality of format information matches.

  According to the present invention, it is possible to search for similar documents based on the format information of the documents. Therefore, it is possible to search for documents that are similar in form in order to refer to or reuse documents created in the past. it can.

  Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. In this example, for example, a program for performing processing corresponding to an electronic computer is installed and realized by using an arithmetic processing function, a storage function, and the like included in the electronic computer.

  FIG. 1 is a block diagram showing a system configuration example when this example is realized by an electronic computer. An overall configuration according to an embodiment of the present example will be described with reference to FIG.

  In this example, an input / output device such as a display device 101, a keyboard 102, and a mouse 103 for a user of the device to perform input / output operations, a calculation unit 104 that executes document search processing, and document data are stored. The document database 105 is stored. The calculation unit 104 includes a document format information extraction unit 107 that extracts format information of the input document 106 input by the user of the apparatus as a search target, and a document and format input based on the extracted document format information 110. A similar-structured document search unit 108 that searches for similar documents, and a search result display unit 109 that edits and displays the searched search result data 111.

  Next, processing of each part of this example will be described.

  First, the user inputs an input document 106 to be searched using an input / output device. The input document may be input by the user from the keyboard 102, or document data created by another device may be input via a storage medium or may be input via a network. Alternatively, a method of selecting from document data registered in the document database 105 may be used.

  The document format information extraction unit 107 analyzes the input document 106 input by the user, and format information of the document, that is, a template used by the document, set properties, document type, and file name. Information such as header and footer, frame and ruled line pattern, document layout, paper size, columnar form, and number of pages is extracted from the information embedded in the document and stored as document format information 110. The property information set in the document includes the title and subtitle attached to the document, the document classification, the document creator, the document size, the document creation date and the update date, and the like. The document type can be acquired as property information, but can also be determined by the extension of the name given to the document when the computer handles the document. For example, if the extension of the name given to the document is “txt”, it is determined that the type of the document is a plain text file. The frame and ruled line pattern and the document layout are determined by analyzing the frame, ruled line, image, and the like included in the input document.

  The similar configuration document search unit 108 evaluates the degree of matching by referring to the document data stored in the document database 105 based on the document format information 110 extracted by the document format information extraction unit 107. For example, when there are a plurality of ruled lines in the input document 106, the degree of matching of the documents in the document database 105 where the ruled lines do not exist is 0%. If half of the parts match, the degree of matching is 50%, and if the positions of the ruled lines match in all parts, the degree of matching is 100%. With regard to the layout of the document format information, for example, when an input document has an image, the document in the document database 105 has a matching degree of a document having no image of 0%, and a document having an image. The degree of matching is 50% if the number of locations where the positions of the images match is 50%, and the degree of matching is 100% if they match at all positions. In this way, the degree to which the extracted document format information matches the document stored in the document database 105 is evaluated, and a document that matches at least one format information is stored as the search result data 111.

  The search result display unit 109 extracts the document searched by the similar configuration document search unit 108 from the search result data 111 and displays it on the display device 101. There are methods for displaying search results, such as displaying in order from the most common types of format information, or displaying the format with the highest degree of match for each format information, which is required by the user. What is necessary is just to edit so that information can be obtained easily.

  FIG. 2 is a block diagram showing a system configuration example according to another embodiment of the present invention. In this example, an input / output device such as a display device 101, a keyboard 102, and a mouse 103 for a user of the device to perform input / output operations, and a document database 105 that stores document data are the first one described above. The configuration is the same as that described in the embodiment. A calculation unit 201 that executes document search processing includes a search condition input unit 202 that inputs document format information as a search condition, a similar configuration document search unit 108 that searches for a document that matches the input search condition 203, and the like. The search result display unit 109 is configured to edit and display the search result data 111.

  In this example, the search condition input unit 202 displays an operation screen for setting a search condition on the display device 101, and the user sets format information as a search condition using the displayed operation screen. The search condition input unit 202 stores the set format information in the search condition data 203. Based on the search condition data 203 set by the search condition input unit 202, the similar configuration document search unit 108 evaluates the degree of matching with reference to the document data stored in the document database 105. The search result display unit 109 extracts the document searched by the similar configuration document search unit 108 from the search result data 111 and displays it on the display device 101. Note that the processing of the similar configuration document search unit 108 and the search result display unit 109 is the same as the processing according to the first embodiment.

  In this way, in this example, even if there is no document to be input as a condition for searching for similar documents, the search condition that has been set is set by setting the formal characteristics of the document that the user wants to search as the search condition. Documents similar to can be searched.

  FIG. 3 shows an example of the document database 105. In the document database 105, an index 301 of a registered document is created and stored together with the document data 302. In the similar document search process, it takes a lot of time to extract format information from all documents registered in the document database each time the target document is searched. It is extracted and registered as the document format index 301. The index includes the registered document format information, the template used by the document, set properties, document type, file name, header and footer, paper size, column format, page Using information such as numbers, this information is extracted and stored as an index in a table format. By registering the format information index in the document database in this way, the similar configuration document search unit 108 uses the document format information 110 extracted from the input document 106 or the search condition data 203 set as a search condition. By referring to the document format index 301 of the document database 105, it is possible to easily search for a document that matches the search condition.

  FIG. 8 is a flowchart showing the processing flow of this example. With reference to FIG. 8, the outline of the overall processing of this example will be described. First, as a search method, it is determined whether to input an input document as a search condition or format information as a search condition (step S801). For example, an operation screen for selecting a search method may be displayed on the display device 101 first, and the user of this device may select which method to search. When an input document is input as a search method, the input document input by the user is analyzed, the document format information is extracted and stored as the document format information 110 (step S802). When inputting a search condition as a search method, an operation screen for setting the search condition is displayed on the display device 101, and the user sets format information as a search condition using the operation screen. Then, the set format information is captured and stored in the search condition data 203 (step S803). Next, the document format information 110 extracted from the input document as the search condition or the search condition data 203 set by the user is retrieved, and a similar document is searched by referring to the document format index 301 registered in the document database. And stored in the search result data 111 (step S804). Finally, the document data is extracted from the search result data 111, edited as the search result, and displayed on the display device 101 (step S805).

  FIG. 9 is a flowchart showing details of the similar configuration document search process. With reference to FIG. 9, a processing example of the similar component document search unit 108 will be described. First, the document format information 110 or search condition data 203 of the input document is extracted according to the search method (step S901). Next, referring to the document format index 301 of the document database 105, a document that matches the format information registered in the document format information 110 or the search condition data 203 is searched and extracted (step S902). Next, the weight value set in the format information is taken out in order to evaluate the overall similarity by adding the weight value added to the format information to the retrieved result (step S903). In order to calculate the similarity based on the extracted weight value, first, one piece of document data is extracted from the search result (step S904). It is determined whether there is document data of the retrieved search result (step S905). If there is a search result, the similarity is calculated based on the weight value of the format information (step S906). If there is no retrieved search result, it means that the calculation of the similarity for all the search results has been completed. Therefore, the retrieved document data and the calculated similarity are combined and stored in the search result data 111 (step S907).

  The similarity calculation process in the similar configuration document search process will be further described. In the similarity calculation process, the total similarity is evaluated by adding the weight value added to the format information to the search result in the similar configuration document search process. FIG. 4 shows an example of the weight value added to the format information. In this example, a weight value is set in advance for each type of document format information used as a search condition. In FIG. 4, each weight value 402 is represented by a numerical value for the format information type 401. This weight value is registered in advance in a storage device in the computer as data used for the similarity calculation process. Further, a means for weighting each format information may be provided so that the user of this apparatus can change the weight value.

  In the similarity calculation process in the similar configuration document search process, the similarity of each searched document is calculated using the weight value of the format information. The similarity is a variable indicating the degree of similarity of documents, and is set so that the value increases as the degree of similarity increases. In the similarity calculation process, first, the initial value of the similarity of each document is set to 0, for example. Next, with respect to the format information of the retrieved document, it is determined which item of the format information that is the search condition matches. Then, all the weight values set in the matching format information are added to the similarity. In this way, the similarity is calculated for each of the search result documents.

  For example, if the “used template” in the format information of the search condition matches, “10” that is the weight value set in the format information is added to the similarity of the document. Similarly, there are three types of format information: “used template (weight value is 10)”, “set property (weight value is 5)”, and “document type (weight value is 20)”. If so, all the corresponding weight values are added and the similarity is 35 points. Also, “file name (weight value is 10)”, “header and footer (weight value is 5)”, “frame and ruled line pattern (weight value is 5)”, and “paper size (weight value is 10)”. In the case of a document in which the four types of format information match, all corresponding weight values are added, and the similarity is 30 points. In these examples, since the weight value for “document type” is relatively high, the former that matches three types of format information is more than the latter that matches four types of format information. Similarity increases.

  As described above, the similar document search unit 108 searches for a similar document from the document database based on the format information, and stores the search result in the search result data 111 together with the similarity. Since the search result display unit 109 can edit the search result display screen based on the similarity by referring to the search result data 111, a document having a high similarity can be displayed at a higher level. Thereby, not only the number of matching format information but also the priority can be given to the search condition, and the document to be searched can be easily obtained.

  FIG. 5 is an example of an operation screen for setting search conditions in the search condition input unit 202. When the method for inputting the search condition is selected as the search method, the search condition input unit 202 displays an operation screen for setting the search condition on the display device 101, and the user uses the displayed operation screen. Set format information as search condition. Here, the processing of the search condition input unit 202 will be described with reference to FIG.

  The operation screen of FIG. 5 includes a format condition selection unit 501 for setting selection and exclusion of a format to be searched, and a search keyword specifying unit 502 for specifying a keyword related to the selected format condition. The format condition selection unit 501 can switch the selection / exclusion by pressing the switch next to each format name with an input device such as a mouse, and selects the format information as a search target or from the search target. Specify whether to exclude. The search keyword specifying unit 502 inputs the contents of the format for searching the target document for the format selected by the format condition selecting unit 501 from a keyboard or the like. For example, when “use template” is selected as the format condition, the template name “style.dot” used as the search keyword is input. If “document type” is selected, “PowerPoint” is input as a search keyword. The search condition input unit 202 stores the format condition and search keyword selected on the operation screen in the search condition data 203, and the similar configuration document search unit 108 refers to the search condition data 203 to select the selected format condition. Only for the search target document in the document database 105.

  When searching for a document, you may want to search by entering the search conditions interactively based on the styles of documents that you have created or viewed in the past. In this example, the search condition format By inputting the above condition as a search keyword, it becomes possible to perform a search using a document style stored in memory.

  In addition, even when a method for inputting an input document serving as a search condition is selected as a search method, not only the format information extracted from the input document is used as it is but also the search condition is selected for the extracted format information. / You may want to exclude or add / modify the format information. In this example, it is possible to arbitrarily set the search condition by providing a search condition setting unit that performs processing for setting the search condition based on the format information extracted from the input document.

  FIG. 7 shows a configuration example when a search condition setting unit is provided. In this example, an input / output device such as a display device 101, a keyboard 102, and a mouse 103 for a user of the device to perform input / output operations, and a document database 105 for storing document data are described above. It is the same as that by the form of. A calculation unit 701 that executes document search processing extracts the format information of the input document 106 that is input as a search target by the user of the apparatus, and stores the document format information extraction unit 107 as document format information 110. A search condition is set based on the information 110, a search condition setting unit 702 that stores the search condition data 203, and a document that matches the set search condition 203 is searched from the document database 105 and stored in the search result data 111. It comprises a similar configuration document search unit 108 and a search result display unit 109 that edits and displays the search result 111. In this configuration, the document format information extraction unit 107, the similar configuration document search unit 108, and the search result display unit 109 are the same as those according to the above-described embodiment.

  FIG. 6 is an example of an operation screen for setting the search condition based on the format information extracted from the input document in the search condition setting unit 702. The search condition setting process in the search condition setting unit 702 will be described with reference to FIG.

  When a method for inputting an input document serving as a search condition is selected as a search method, the document format information extraction unit 107 extracts format information from the input document. However, instead of using the format information as it is, the search condition setting unit 702 does not use it. The search condition is set using the extracted format information. The search condition setting unit 702 displays an operation screen for setting search conditions as shown in FIG. The operation screen in FIG. 6 includes a format condition selection unit 601 for setting selection and exclusion of a format to be searched, a format information display unit 602 for displaying format information extracted from the input document, and a search keyword for specifying a keyword related to the format condition. It consists of a designation unit 603. The format information of the input document extracted by the document format information extraction unit 107 is displayed on the format information display unit 602 of the input document. The user of this apparatus refers to the displayed format information and uses the format condition selection unit 601 to specify whether the format is selected as a search target or excluded from the search target. The selection / exclusion designation method is the same as that of the format condition selection unit 501 in FIG. For example, if you do not want to search for templates used in a document, set the switch next to “Use template” to the excluded state (the selection switch in the figure is white), and select “Property” Switches such as “document type”, “file name”, and “header” are selected (the selection switch in the figure is black).

  If the user wants to search under conditions different from the format information extracted, the keyword to be searched is specified in the search keyword specifying unit 603. When a search keyword is specified, the content of the format specified here is used as a search condition instead of not using the format information extracted from the input document. For example, in the property information extracted from the input document, when the document creator is “XXX” and the document creator to be actually searched is “YYY”, the search keyword portion in the property column Enter "YYY" in If the header information is not extracted from the input document and the header of the document to be actually searched is “patent”, “patent” is input in the search keyword portion of the header column. As described above, when a keyword is specified in the search keyword specifying unit 603, the search condition can be searched with the content specified as the keyword, and the other search conditions can be searched with the same content as the input document.

  The search condition setting unit 702 stores the format conditions and search keywords set on this operation screen as the search condition data 203, and the similar configuration document search unit 108 does not use the format information extracted by the document format information extraction unit 107. With reference to the search condition data 203 set by the user, the document database 105 is checked for a match.

  Thus, by providing the search condition setting unit 702, the format information extracted from the input document can be used as the search condition as it is, and the search condition can be arbitrarily changed with respect to the extracted format information. It is possible to search by a document style stored in the user's memory while referring to the style of a specific document. Also, in setting search conditions, it is possible to search without changing the format conditions extracted from the input document in advance. It can be performed.

  Next, still another embodiment of the present invention will be described. FIG. 10 shows an example in which a document is registered in the database with the format information of the input document as a category and is searched. This example includes a calculation unit 10 that performs document registration, a calculation unit 20 that performs document search, and a document database 30. The document registration calculation unit 10 includes a document format information extraction unit 107 that extracts format information of a document to be registered in the database, and a document registration unit 11 that classifies the documents based on the extracted format information and registers them in the database. . The document search calculation unit 20 includes a document category input unit 21 that specifies a document category to be searched, a document format information extraction unit 107 that extracts format information of an input document as a search condition, and a document in a specified category. A similar document search unit 22 for searching for documents having similar formats, and a search result display unit 109 for editing and displaying the search results. In FIG. 10, a display device for a user of this apparatus to perform input / output operations, and an input / output device such as a keyboard and a mouse are not shown, but are configured in the same manner as in the other embodiments. In this example, the processing of the document format information extraction unit 107 and the search result display unit 109 is the same as that according to the embodiment described above.

  First, the processing of the document registration calculation unit 10 will be described. When a new document is registered in the document database, first, the document 40 to be registered is input. The document 40 to be registered may be input by a user from a keyboard, or document data created by another device may be input via a storage medium or may be input via a network. Next, the document format information extraction unit 107 extracts the format information of the input registered document 40 as a category for registration in the document database 30. Format information that can be used as a category includes a template used by a document, set properties, document type, file name, header and footer, paper size, columnar form, number of pages, and the like. Here, the document properties will be described as an example. The property information set in the document includes the title and subtitle attached to the document, the document classification, the document creator, the document size, the document creation date and the update date, and the like. The document format information extraction unit 107 extracts properties set in the input registered document 40. Here, description will be made by taking as an example the document classification in the property (the document classification input by the creator when the document is stored). For example, it is assumed that there are classifications such as “report”, “technical document”, “XX request form”, etc. as the classification of documents in the property. The document format information extraction unit 107 extracts the document classification from the registered document 40, and the document registration unit 11 performs “report”, “technical document”, “XX request form”, etc. based on the classification extracted from the document. The registered document 40 is registered in the database after classification into any of the above. For example, when the document classification of the registered document 40 is “report”, the document registration unit 11 registers the registered document 40 in the “report” category of the document database. This eliminates the need for an interactive operation in which the user is conscious of document classification and assigns the folder.

  Next, when a document is searched from the document database classified and registered in this way, the user first inputs the document category at the document category input unit 21 of the document search calculation unit 20. Next, search conditions are set. Here, a case where an input document as a search condition is input as a search method will be described as an example, but another search condition setting method may be used. When an input document is input as a search condition, the document format information extraction unit 107 extracts format information of the input document. Then, based on the extracted format information, the similar document search unit 22 searches for a similar document from the document database 30. Here, the similar configuration document search unit 22 first narrows down the search target of the document database 30 based on the category input by the category input unit 21. Then, similar documents are searched for the narrowed-down range of documents, and the search result display unit 109 edits and displays the search results. For example, when the user specifies the document classification “technical data” in the property as the document category to be searched, the similar-structured document search unit 22 stores the document data registered in the “technical data” category of the document database 30. Search for similar configuration documents.

  In this way, by registering a document in the database with the format information of the document as a category, the target document can be searched quickly. In addition, by narrowing down the search target using categories, it is possible to prevent unnecessary documents from being included in the search results and improve the document search accuracy, resulting in shortening the time for users to search for documents. In addition, the search effort can be reduced.

It is a block diagram which shows the structural example by one embodiment of this invention. It is a block diagram which shows the structural example by other embodiment of this invention. It is explanatory drawing which shows the example of a document database by one embodiment of this invention. It is explanatory drawing which shows the example of the weight value of the format information by one embodiment of this invention. It is explanatory drawing which shows the example of an operation screen (1) by one embodiment of this invention. It is explanatory drawing which shows the operation screen example (2) by one embodiment of this invention. It is a block diagram which shows the structural example by other embodiment of this invention. It is a flowchart which shows the example of a search process by one embodiment of this invention. It is a flowchart which shows the similar structural document search processing example by one embodiment of this invention. It is a block diagram which shows the structural example by other embodiment of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 101 ... Display apparatus, 102 ... Keyboard, 103 ... Mouse, 104 ... Operation part, 105 ... Document database, 106 ... Input document, 107 ... Document format information extraction part, 108 ... Similar structure document search part, 109 ... Search result display part 110 ... Document format information, 111 ... Search result data, 201 ... Calculation unit, 202 ... Search condition input unit, 203 ... Search condition data, 301 ... Document format index, 302 ... Document data, 401 ... Type of format information, 402 ... Weight value, 701 ... Calculation unit, 702 ... Search condition setting unit, 10 ... Document registration calculation unit, 20 ... Document search calculation unit, 30 ... Document database, 40 ... Registered document, 11 ... Document registration unit, 21 ... Document Category input part, 22 ... Similar composition document search part

Claims (10)

In a document search apparatus using a computer including an input / output device, a calculation unit that performs processing, and a storage device that stores data,
A document format information extraction unit for extracting the format information of the document;
A similar configuration document search unit that searches a document database stored in a storage device for documents similar to the format information extracted by the document format information extraction unit;
A document search device comprising a search result display unit for outputting a search result to an input / output device. In a document search apparatus using a computer including an input / output device, a calculation unit that performs processing, and a storage device that stores data,
A search condition input unit for inputting document format information as a search condition;
A similar document search unit that searches a document database stored in the storage device for documents similar to the format information input by the search condition input unit;
A document search device comprising a search result display unit for outputting a search result to an input / output device. The document search apparatus according to claim 1, wherein
Means for adding weight values to the format information;
When the format information extracted by the document format information extraction unit and the format information of the document stored in the document database match, the weight value added to the format information is added to the similarity of the document. Means for calculating the similarity of
A document search apparatus that displays search results according to the degree of similarity. The document search device according to claim 2, wherein
Means for adding weight values to the format information;
When the format information input by the search condition input unit matches the format information of the document stored in the document database, the weight value added to the format information is added to the similarity of the document to Provide a means to calculate the similarity,
A document search apparatus that displays search results according to the degree of similarity. The document search apparatus according to claim 1, wherein
Based on the format information extracted by the document format information extraction unit, means for selecting and excluding the format to be searched,
A means for changing the contents of the extracted format information and specifying a keyword related to the format condition to be searched;
A document search device configured so that search conditions can be arbitrarily set. The document search device according to any one of claims 1 to 5,
When registering a document in the document database, a document registration processing unit is provided for extracting the format information of the document, classifying the extracted format information into a document database as a category, and registering it.
In document search, a document search apparatus characterized by providing means for specifying a document category to be searched and narrowing down a search range of a document database. In a document search method for searching a document from a document database based on a document input by a user,
A document format information extraction process for analyzing the input document and extracting the format information of the document;
Similar document search processing for searching a document having format information that matches the extracted format information with reference to the document database based on the extracted format information;
A document search method comprising: a search result display process for editing and outputting a searched document. In a document retrieval method for retrieving a document from a document database,
Search condition input processing for the user to set document format information as a search condition;
Similar document search processing for searching a document having format information that matches the input format information with reference to the document database based on the input format information;
A document search method comprising: a search result display process for editing and outputting a searched document. The document search method according to claim 7, wherein
Provide a process to add weight values to the format information,
For the document searched by the similar document search process, when the format information extracted by the document format information extraction process matches the format information of the document stored in the document database, the weight added to the format information Provide a process to calculate the similarity by adding the value to the similarity of the document,
A document search method that displays search results according to the degree of similarity. The document search method according to claim 8.
Provide a process to add weight values to the format information,
For documents searched by similar-structured document search processing, if the format information input by the search condition input processing matches the format information of the document stored in the document database, the weight value added to the format information To calculate the similarity by adding to the similarity of the document,
A document search method that displays search results according to the degree of similarity.

Images