JP4615279B2 - Document classification recognition apparatus, method thereof, document classification apparatus, and program thereof - Google Patents

Description

  The present invention records a document classification recognition device that recognizes a document classification using a plurality of document information associated with classification information related to the classification, a method thereof, a document classification device, a program thereof, and a program thereof. The present invention relates to a recording medium.

  Conventionally, various configurations have been known as devices for classifying documents (see, for example, Patent Document 1).

  The document described in Patent Document 1 stores an input document in a classification document database for each classification based on the classification rules stored in the classification rule database. A keyword is extracted for each classification from the documents stored in the classified document database and stored in the keyword database. If the documents stored in the classification document database are stored in the classification creation condition database and satisfy the classification creation conditions entered by the user for the first time, the keywords stored in the keyword database are used and stored in the classification creation rule database. A new classification rule for classifying documents according to the classified classification creation rule is created. When a new classification rule is created, the documents stored in the document classification database are reclassified. In this way, when the user first inputs a classification creation rule, a classification rule is created to classify the document, and the input document is automatically classified based on this classification rule. It has been.

  The method described in Patent Document 2 calculates the frequency of classification into predetermined classifications in all classifications for each classification item by using adjacent word strings included in documents classified in advance as classification items. The probability of classifying is calculated and accumulated. Then, when a document to be classified is input, a word string is extracted from the classification item using words in the document, and the extracted word string is matched with a continuous word string stored in advance. A classification probability belonging to a certain classification is extracted. Using the classification probabilities of this classification item, the probability that a document is classified for a certain classification is calculated individually and displayed as the classification result of the document in order from the highest probability of this fraction probability. It is taken.

  In the document described in Patent Document 3, a word that appears in only one classification in the document data is detected from the classified document data, and is registered in the classification dictionary as a keyword representing the classification. Words in uncategorized document data are detected, the number of matches with keywords registered in the classification dictionary is detected, and the classification with the largest number of matches among the matched classifications is set as the classification result.

  The one described in Patent Document 4 selects effective words useful for classification from the words included in the learning document, and stores them as effective word dictionaries by vector expression based on the appearance frequency and other co-occurrence situations for each validity. . For the learning document and the classification target document, a paragraph vector is obtained by referring to the effective word dictionary, and the other topic paragraph is determined from the distribution of the paragraph vector. An effective paragraph vector is extracted from the paragraph vectors with reference to other topic paragraphs, and a document vector is obtained. A folder vector of each category is obtained using the document vector of the learning document. A configuration is adopted in which the category to which the classification target document belongs is determined according to the comparison result between the document vector of the classification target document and the folder vector of each category.

According to the method described in Patent Document 5, an evaluation function based on weighting obtained from the accuracy rate for a classification performed in the past is stored in a database for each person in charge of classification, and a typical document characterizing the classification for each classification in advance. Have it ready. The manual classification is scored for each classification from the classification result of the target document by the classifier and the evaluation function. The degree of similarity between the target document and the typical document is calculated, and the automatic classification is scored for each classification. The classification with the highest total of the manual classification score and the automatic classification score for each classification is used as the final classification result, and the manual classification and automatic classification are merged to obtain an accurate classification result. .

JP 11-203302 A (page 5 left column to page 7 right column) JP-A-8-287097 (page 7, left column to page 9, right column) JP-A-6-348755 (right column on page 3 to right column on page 6) JP-A-11-85796 (page 6, right column to page 9, left column) Japanese Patent Laid-Open No. 11-45247 (page 3, left column to page 6, right column)

As described above, in the conventional configuration as described in Patent Document 1 to Patent Document 5, a word set such as a keyword characterizing a predefined classification item, a feature pattern, a rule, and the like are set in advance, and the classification item Is used to collate a classification item with a document to be classified and determine a classification item. Thus, as appropriate classification is obtained, it is necessary to set rough steering Me a predetermined classification item, a problem device construction is complicated.

An object of the present invention, in view of the above problems, recognition easy document classification recognizer classification document, the method, Document classification device, and to provide their program.

Document classification recognizer of the present invention, the words a recognized document classification recognizer classification including plain coarse Document information, and acquisition means for acquiring a word included in the document information of the recognition target of the classification, and words included in the document information and the the classification and configured document information stored in the document storage means stores a plurality associated comprise a plurality of words, the word acquired by the acquisition means The judgment is made from judgment means for judging identity, literature selection means for selecting the literature information stored in the literature storage means based on the identity, and the classification associated with the selected literature information. A class acquisition unit that calculates a score based on the identity and selects candidates for the class based on the total score for each class, and the determination unit matches the acquired word Document memory About words in the document information stored, as the matching word count the number of occurrences in the words that the said document storing means is stored in the document information and the recognition target document information that matches the document storage The total number of words constituting the document information stored in the means and the document information to be recognized is set as the total number of words, and the total number of words is divided by the number of matched words to obtain the weight information as the identity. The classification acquisition means calculates the score based on the weight information calculated for the word.

In the present invention, taken acquires the words included in the recognition target document information classified at yield means stores plural literature storage means the classification and are associated with the configured document information including a plurality of words The determination means determines the identity between the word included in the document information stored in and the acquired word . As for this identity, for words in the document information stored in the document storage means that match the acquired word, the matching words are among the document information stored in the document storage means and the document information to be recognized. the number that appears as the matching word count, the total number of words constituting the document information of the recognition target document information stored in the document storage means and the total number of words, dividing the total number of words in the word-match count Thus, the weight degree information as the identity is calculated. The document information stored in the document storage means based on the determination the identity elected in the literature selecting means, by the classification acquiring means a classification associated with the document information based on the scores calculated based on the weight degree information Acquired as classification of recognition target. In this configuration, the document information of the contents similar to the recognition target document information classification, the identity of the determination of the word in advance classification is included in the word and document information of the recognition target included in the document information associated recognized, since recognizing the classification of document information similar content as the classification of the document information of the recognition target, for example, using a database of document information existing classification is recognized, you are easy operations classification Bibliography Can be easily recognized.

  In the present invention, the determination means associates the score information related to the score corresponding to the identity with the document information in the document storage means, and the document selection means selects the document information having a high score in the score information. It is preferable to adopt a configuration to In this configuration, since the score information relating to the score corresponding to the identity determined by the determination means is associated with the document information of the document storage means, and the document information associated with the high score information is selected by the document selection means, The degree of similarity of the document information whose contents are similar to the document to be recognized is easily recognized by the score, the identification of the document information having similar contents is obtained by a simple calculation, and the classification is easily recognized.

Furthermore, in the present invention, the classification acquisition unit, the acquired associate classified as both the score information associated with the document information that has been selected in the literature selecting means, said classification every acquired from the document information It is preferable that the total score is calculated by adding the scores of the score information, and the classification having a high total score is acquired as the classification of the recognition target. In this configuration, by the classification acquisition unit acquires score information classification Contact and in document information selected at document selecting means calculates the total score by summing the scores obtained classified every score information, a high total score Since the classification is acquired as the classification of the recognition target, the classification corresponding to the content similarity situation is appropriately selected.

And in this invention, the said classification acquisition means calculates the appearance frequency for every classification | category linked | related with the said literature information selected by the said literature selection means, The classification | category with high appearance frequency of each of these classification | category is the said recognition object. It is preferable that the configuration is obtained as a classification of In this configuration, the classifying acquisition means calculates the appearance frequency of each classification associated with the selected the document information in the literature selection means, the frequency of occurrence of each category is a high classifying the classification of recognized, contents A suitable classification corresponding to can be recognized.

Furthermore, in the present invention, the classification acquisition unit is preferably configured for calculating a counted by frequency classification of each document information selected by said document selecting means for each classification. In this configuration, the classifying acquisition means, because it calculates the count to frequency for each classification a classification of each document information selected in the literature selecting means, by a suitable classification simple operation corresponding to more content It is easily recognized.

In the present invention, the classification, it is preferable that the structure is the International Patent Classification. In this configuration, since the use of the International Patent Classification it is internationally unified content classification as classified, along with a unified classification is obtained, which is an existing patent documents International Patent Classification is associated patents A database such as public gazette data can be used, and it is not necessary to construct a document storage means in advance, and classification can be easily recognized with a simple configuration.

  Furthermore, in the present invention, it is preferable to have a configuration including a display control means for displaying the classification acquired by the classification acquisition means on the display means so as to be selectable by an input operation. In this configuration, the display control means causes the display means to display the classification acquired by the classification acquisition means so as to be selectable by an input operation. For this reason, the user can confirm the classification recognized by the calculation, and it is easy to make appropriate settings and corrections, and a more appropriate classification setting can be obtained.

  In the present invention, it is preferable that the display control unit is configured to cause the display unit to display explanatory information regarding the contents of the classification when recognizing that the classification displayed on the display unit is selected by an input operation. . In this configuration, when the display control means recognizes that the classification displayed on the display means has been selected by the input operation, the display means displays the explanation information on the contents of the classification, so that the recognized classification contents are used. Can be easily confirmed by a user, and usability is improved.

  In the present invention, it is preferable that the display control unit performs control to display the classification acquired by the classification acquisition unit in a display form corresponding to the identity. In this configuration, the display control means causes the classification acquired by the classification acquisition means to be displayed in a display form corresponding to the identity, for example, a color or brightness in a form corresponding to the identity. Can be easily recognized by the user, and is particularly effective when, for example, a plurality of classifications are displayed.

  Further, in the present invention, it is preferable that the display control unit performs control to display the classification acquired by the classification acquisition unit together with the score information related to the score corresponding to the identity. In this configuration, since the display control means displays the classification acquired by the classification acquisition means together with the score information related to the score corresponding to the identity, it is easy for the user to match the content of the document information to be recognized. This is particularly effective when a plurality of classifications are displayed, for example.

In the present invention, when displaying on the display means a word acquired by the acquisition unit, including the display control means for displaying on the display means the document information in the display mode different from the case of displaying the other words It is preferable to adopt the configuration described above. In this configuration, the display control unit, collected since the word acquired in resulting unit is displayed on the display means document information in a display form different from the other word, the classification object recognition literature information and same as that document information The user can easily recognize words that cause classification recognition in the literature information of classification, and usability is improved.

Furthermore, in the present invention, the display control means preferably the words displayed on the different display form and the other word is selected by an input operation and structure to be displayed on the display means the classification obtained by the word . In this configuration, the display control means, so to be displayed on the display means when selected classification obtained by the word by the displayed word is input operations to a different display form other words, confirmation of the classification of the document information is It becomes easy and usability improves.

In the present invention, at least one of changing the document information transmitted from the server device connected via the network and stored in the document storage means and changing the storage state of the document information. A storage control unit configured to control the document storage unit when the operation information related to the data distribution operation including one is recognized and to change the document information and the storage state in response to the data distribution operation; It is preferable to adopt the configuration described above. In this configuration, with respect to documents storage means of the table structure that stores a plurality of document information that has been built into a single data structure classification is associated, to deliver data from a server device which is connected to the network, the stored contents The classification system support capability can be expanded / changed by performing data addition / change or new writing and data distribution operation via the network.

The document classification apparatus of the present invention, a document information includes a plurality of words, the document storage means classification and are more stored in association, a given document information using the document storage means The document classification recognition apparatus according to the present invention for recognizing the classification of the document is provided.

According to the present invention, a document storage means in which a plurality of document information including a plurality of words and a plurality of classifications are associated and stored is used to classify predetermined document information into a document classification described in the present invention. Recognize by recognition device. Thus, for example, existing classification can utilize a database of document information has been recognized, can be easily recognized by the classification easy operation document information.

  In the present invention, it is preferable to include a storage control unit that associates the classification recognized by the document classification recognition apparatus with the document information to be recognized and stores the document information in the document storage unit. In this configuration, the storage control means associates the classification recognized by the literature classification recognition apparatus with the literature information to be recognized, and this literature information is stored in the literature storage means, so that the literature information is automatically accumulated and is easy to use. improves. In addition, for example, it is easy to construct a database that is easy to use, such as constructing document information for each classification.

Further, document classification recognition method of the present invention, a computer, word a sort recognizes Document classification recognition method including plain coarse Document information, the computer is included in the recognition the document information of the target of the classification Upon obtaining the word, and words included in the document information and the the classification and configured document information stored in the document storage means stores a plurality associated comprise a plurality of words, and words that the obtained Judge identity. During this identity determination, the word in the acquired document information stored in said document storage means coincides with the word, the recognition target document information word that the match is stored in the document storage means of the word-match count the number of occurrences in the document information, the total number of words constituting said literature document information of document information stored in the storage means and the recognition target and the total number of words, the total number of words Is divided by the number of matching words to calculate weight degree information as the identity. The document information stored in the document storage means based on the determination the identity out selection, from the classification associated with the selected the document information, a score based on the weight of information as identity which is the arithmetic A candidate for the classification is selected based on the total score for each classification.

In the present invention, the operation of the document classification recognition apparatus according to claim 1 is developed by a computer- implemented method, and the same effects as those of the invention according to claim 1 are achieved.

The document classification program of the present invention, computer, document classification recognition apparatus according to the present invention, or, wherein the function as document classification apparatus according to the present invention.

In the present invention, for example, by installing to take advantage of the general-purpose computer, it is possible to be implemented to function as a document classification apparatus according to the literature classification recognizer, or the present invention according to the present invention, the present invention Can be promoted significantly.

A document classification program according to the present invention causes a computer to execute the document classification recognition method according to the present invention .

In the present invention, for example, by installing to take advantage of the general-purpose computer, since the document classification recognition method according to the present invention is executed by a computer, it promotes use of the present invention can be achieved significantly.

In the present invention, the computer includes, for example, one personal computer, a configuration in which a plurality of computers are combined in a network, an element such as a microcomputer, or a circuit board on which a plurality of electronic components are mounted.

[Configuration of document information classification system]
Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In this embodiment, the document classification recognition device of the present invention is provided, and the international patent classification, which is a classification of predetermined document information, is recognized using document information relating to industrial property rights, utility models, and designs. Although the configuration of the document information classification system as the document classification apparatus is exemplified, any document information related to other industrial property rights such as trademarks and technical documents can be targeted. In addition, the classification is not limited to the international patent classification, for example, literature information such as preset contents such as “report”, “sport”, “entertainment”, “medicine”, “math”, “biology” Any classification that identifies the contents of FIG. 1 is a conceptual diagram showing a schematic configuration of a document information classification system in the present embodiment. FIG. 2 is a block diagram showing a schematic configuration of a terminal device constituting the literature information classification system. FIG. 3 is a block diagram showing a schematic configuration of a server device constituting the literature information classification system. FIG. 4 is a schematic diagram conceptually showing the table structure of the document information storage area. FIG. 5 is a schematic diagram conceptually showing the table structure of the document-related classification storage area.

  In FIG. 1, reference numeral 100 denotes a document information classification system as a document classification device. This document information classification system 100 is for recognizing classification of document information related to documents related to industrial property rights such as patents, utility models, and designs. System. The document information classification system 100 includes a terminal device 200 and a server device 300 that also functions as a document classification device to which the terminal device 200 is connected via a network 110 so that various types of information can be transmitted and received.

  Here, the network 110 connects the server device 300 and the terminal device 200 so that various data can be transmitted and received. The network 110 includes an Internet based on a general-purpose protocol such as TCP / IP (Transmission Control Protocol / Internet Protocol), an intranet such as a LAN (Local Area Network), and a plurality of base stations capable of transmitting and receiving information via a wireless medium. Any configuration that transmits / receives data, such as a network such as a communication line network or a broadcast network, or a wireless medium itself that directly transmits / receives data, can be used. Here, any medium such as radio waves, light, sound waves, and electromagnetic waves can be applied as the wireless medium.

  The terminal device 200 is a device for requesting the server device 300 to recognize a document classification, and examples thereof include a personal computer and an office computer. Then, as shown in FIG. 2, for example, the terminal device 200 includes a terminal communication unit 210, a terminal input unit 220 as an operation unit, a terminal display unit 230 as a display unit, a terminal storage unit 240, and a terminal processing unit. 250.

  The terminal communication unit 210 is connected to the server device 300 via the network 110. The terminal communication unit 210 can receive a terminal signal from the server device 300 via the network 110, performs input interface processing set in advance by acquiring the terminal signal, and sends it to the terminal processing unit 250 as a processing terminal signal. Output. Further, the terminal communication unit 210 can receive a processing terminal signal from the terminal processing unit 250, performs an output interface process set in advance by obtaining the input processing terminal signal, and transmits it as a terminal signal via the network 110. To the server device 300.

  The terminal input unit 220 includes various operation buttons and operation knobs (not shown) that are input by a user who requests execution of classification recognition processing using, for example, a keyboard or a mouse. As input operations such as these operation buttons and operation knobs, in addition to setting operation contents of the terminal device 200, input settings for lexical information of literature information, various condition items for classification recognition processing in the server device 300, and the server device 300. This includes setting items such as determination, correction, and change of the classification selected in, and a request for obtaining literature information. Then, the terminal input unit 220 appropriately outputs and sets a predetermined signal corresponding to the input operation to the terminal processing unit 250. The terminal input unit 220 is not limited to an input operation such as an operation button or an operation knob. For example, an input operation using a touch panel provided on the terminal display unit 230, an input operation using voice, or the like, or optical character reading. Any configuration in which various input items can be set and input, such as reading with a device (Optical Character Reader: OCR) or an image reading device, can be applied.

  The terminal display unit 230 is controlled by the terminal processing unit 250 and displays the image data from the terminal processing unit 250 on the screen. The image data includes, for example, a screen display that prompts input of various condition items for classification recognition processing, and a classification selected by the server device 300 is selected or determined by an input operation by the terminal input unit 220, and further corrected or changed. Various image data such as screen display that can be displayed, screen display that displays a list of specific information for identifying document information with similar classifications, and operation data of the terminal device 200 and image data related to other application software Can be displayed. Examples of the terminal display means 230 include various displays such as a liquid crystal display panel, an organic EL (Electro Luminescence) panel, a PDP (Plasma Display Panel), a CRT (Cathode-Ray Tube), an FED (Field Emission Display), Examples thereof include an electrophoretic display panel.

  The terminal storage unit 240 appropriately stores various types of information acquired from the server device 300 via the network 110 and input items input and operated by the terminal input unit 220 as well as various types of information acquired from the outside. The terminal storage unit 240 also stores various programs developed on an OS (Operating System) that controls the operation of the entire terminal device 200. The terminal storage unit 240 includes any configuration capable of storing various types of information, such as a memory, in addition to a configuration including a drive and a driver that are readable and stored in a recording medium such as an HD (Hard Disc) or an optical disc. It can be used, and it is good also as a structure provided with both the drive and the memory, and more than one.

  The terminal processing unit 250 includes, for example, a CPU (Central Processing Unit), and various input / output ports (not shown), for example, a communication control port to which the terminal communication unit 210 is connected, a key input port to which the terminal input unit 220 is connected, and a terminal display A display unit control port to which the unit 230 is connected; a print control port to which a printing apparatus as an output unit (not shown) that performs print output can be connected; a storage port to which the terminal storage unit 240 is connected; The terminal processing unit 250 includes, as various programs, a lexical setting unit 251, a request setting unit 252, a result acquisition unit 253, an output control unit 254 that also functions as a display control unit, and the like.

  The lexical setting unit 251 sets a lexical word for recognizing the classification of the document information, for example, a word or phrase included in the document, based on the information input by the input operation by the terminal input unit 220. Specifically, lexical information relating to words and phrases input by the terminal input means 220 by a predetermined input operation is generated. As lexical information, for example, Japanese such as hiragana, katakana, kanji, etc., converted according to input such as so-called Roman alphabet input by an input program (IME) as input support means. Not only words but also any words, phrases, or parts of words, such as English words input in alphabet, Chinese, and Hangul characters. In addition, the lexical information selects not only words and phrases directly input by the input operation using the keyboard in the terminal input means 220 but also some words and phrases in the document displayed on the terminal display means 230. It is generated based on the words and phrases input by the input operation.

Based on the lexical information of the lexical information acquired by the lexical setting unit 251, the request setting unit 252 generates request information for requesting the server apparatus 300 to execute processing for recognizing the classification of the document information from which this lexical is extracted. . That is, the request information, lexical and, the condition information is a condition recognition of this classification, and the terminal identification information having such customer information for specifying a user or the terminal device 200 that the request for processing, 1 Associated as one data structure. Then, the request setting unit 252 performs processing for transmitting the generated request information to the server apparatus 300 via the network 110 by the terminal communication unit 210.

  The result acquisition unit 253 acquires the result information of the classification recognition process acquired from the server device 300 by the terminal communication unit 210 via the network 110. The acquired result information is appropriately stored in the terminal storage unit 240 or a memory (not shown). The result information includes, for example, an international patent classification that is a classification, score information related to a score that is the probability of the classification, and document specific information that identifies document information associated with the classification information related to the classification, such as an application number and a registration number. In addition to the above, information on the document information itself is also included.

  The output control means 254 appropriately processes the result information acquired by the result acquisition means 253 based on the output status set by the input operation by the terminal input means 220 and displays the result information on the terminal display means 230 or is not shown. Controls the output of the result information that is printed out by the connected printing device. Further, the output control unit 254 controls the terminal display unit 230 to display a plurality of classification candidates output from the server device 300 so as to be selectable by an input operation by the terminal input unit 220.

  The server apparatus 300 is an apparatus that selects the international patent classification that is the classification of the requested document information including the lexical phrase set and input from the terminal apparatus 200 and provides the selected selection to the terminal apparatus 200. Various information can be transmitted to and received from the apparatus 200. For example, as shown in FIG. 3, the server apparatus 300 includes a server communication unit 310, a server input unit 320, a server display unit 330, a server storage unit 340 that functions as a document storage unit, and a document classification as a calculation unit. And a server processing unit 350 as a recognition device.

  The server communication unit 310 performs a preset input interface process on a server signal input via the network 110 and outputs the processed server signal to the server processing unit 350 as a processing server signal. Further, when a processing server signal to be transmitted from the server processing unit 350 to the terminal device 200 is input from the server processing unit 350, the server communication unit 310 executes an output interface process set in advance for the input processing server signal. And it outputs to the terminal device 200 via the network 110 as a server signal. The server signal may be output only to a predetermined terminal device 200 as appropriate based on information described in the processing server signal, or may be output to all the terminal devices 200 connected via the network 110. Is possible.

  The server input unit 320 includes various operation buttons and operation knobs (not shown) that are input by a server administrator or the like, for example, like the terminal input unit 220. These input operations such as operation buttons and operation knobs are various settings such as setting of operation contents of the server device 300, setting input of information stored in the server storage unit 340, updating of information stored in the server storage unit 340, and the like. It is matter. Then, the server input unit 320 appropriately outputs a signal corresponding to the setting item to the server processing unit 350 by the setting item input operation to input the setting. Note that the input operation is not limited to a configuration including operation buttons, operation knobs, and the like, as with the terminal input unit 220.

  As the server display unit 330, various display devices are used, for example, similarly to the terminal display unit 230. The server display unit 330 is connected to the server processing unit 350 and appropriately displays image data output from the server processing unit 350 under the control of the server processing unit 350.

  Similar to the terminal storage unit 240, the server storage unit 340 uses any configuration capable of storing various types of information, and includes a document information storage region 341 as a document storage unit, a document-related classification storage region 342, and the like. ing.

  The document information storage area 341 has a table structure for storing a plurality of document information 341A. For example, as shown in FIG. 4, the document information 341A is configured by associating text information 341A1, identification information 341A2, and the like as one data structure. The body information 341A1 includes a plurality of pieces of text information in a text format, such as sentences and phrases composed of a plurality of words. The document information 341A may include drawing information in an image data format. The identification information 341A2 is, for example, an ID (identification) number that is document-specific information that identifies the document information 341A. The document information 341A is appropriately associated with an application number, a publication number, supplementary information that is bibliographic information such as an applicant and an author, and the like.

For example, as shown in FIG. 5, the document-related classification storage area 342 includes a document classification in which the identification information 341A2 of the document information 341A and the classification information 342A1 related to the international patent classification that is the classification of the document information 341A are used as one record. A plurality of correspondence information 342A is stored, that is, a table structure in which a plurality of correspondence information 342A is recorded for every document information 341A stored in the document information storage area 341 is constructed. In addition, when a database having an international patent classification is constructed as the bibliographic information of the document information 341A stored in the document information storage region 341 in advance, the document-related classification storage area 342 is identification information of each document information 341A. associated extracts and 341A2 and classification information 342A1 of the International Patent classification, it automatically generates a document classification corresponding information 342A by multiple recording, is easily constructed.

  The server storage unit 340 also stores various programs developed on the OS that controls the operation of the entire server device 300 and the entire document information classification system 100. Further, the server storage unit 340 can appropriately store various information received from the terminal device 200 and the like and various information input by the server input unit 320.

  The server processing unit 350 includes a CPU similar to the terminal processing unit 250, and includes various input / output ports (not shown) such as a communication control port to which the server communication unit 310 is connected, an input port to which the server input unit 320 is connected, and a server display. A display control port to which the means 330 is connected, a storage port to which the server storage means 340 is connected, and the like. Then, as shown in FIG. 3, the server processing unit 350 includes, as various programs stored in the server storage unit 340, a request information acquisition unit 351 that also functions as a word information acquisition unit, a determination unit 352, and a document selection unit. 353, classification acquisition means 354, notification means 355, search means 356, billing calculation means 357, settlement means 358, storage control means 359, and the like.

  The request information acquisition unit 351 acquires request information transmitted from the terminal device 200 via the server communication unit 310. Then, the request information acquisition unit 351 acquires the lexical information of the lexical information included in the request information and outputs it to the determination unit 352.

  The determination unit 352 recognizes the lexical phrase output from the request information acquisition unit 351 and determines the identity of this lexical phrase and the lexical phrase included in the document information 341A stored in the document information storage area 341 of the server storage unit 340. . For example, in the case of a single word as a lexical phrase, the number of documents containing the same word in the document information 341A is recognized, and the document information 341A having a larger number is determined to have higher identity. Further, when there are a plurality of words as a lexical word, the ratio and the number of the same words as the words included in the document information 341A are recognized, and the document information 341A having a higher ratio has a higher identity and further the number thereof. It is judged that the greater the number, the higher the identity. Further, when the phrase is a phrase, the number of the same phrase included in the document information 341A, and the ratio and the number of the same word when the phrase is decomposed into a plurality of words by morphological analysis or the like are recognized. Then, it is determined that the document information 341 </ b> A with more identical phrases has higher identity, and the greater the ratio and the number of words that are divided into a plurality of words, the higher the identity. In the case of a phrase, not only the ratio of the same word when it is decomposed into a plurality of words but also the identity in a phrase obtained by further dividing the phrase may be determined. Then, the determination unit 352 outputs the identity information regarding the identity for each document information 341 </ b> A to the document selection unit 353.

  In addition, as a specific calculation of identity, for example, weight information on the identity between a lexical word and the text information 341A1 of the document information 341A is calculated, and the similarity of the text information 341A1 to the lexical based on the weight information. Similarity is calculated as a score related to the state, and is regarded as the same. First, as the calculation of the weight information, for example, the weight of the body information 341A1 that matches each word of the lexical word is calculated. That is, if a plurality of words as lexical characters are q1,..., Qn (n is the number of words), and each word of the text information 341A1 is d1,..., Dm (m is the number of words), the weight degree W (qi, dj, q, d) is calculated by the following equation 1 for qi = dj. The calculated weight W (qi, dj, q, d) is appropriately stored in the server storage unit 340.

[Equation 1]
W (qi, dj, q, d) = Log (total number of words / (qi = number of words in which dj appears))
× (Number of occurrences of qi lexical word)
X (number of appearances of word of dj body information 341A1)

  Based on the calculated weight, the similarity of the text information 341A1 with respect to the lexical word is calculated to determine the identity. For example, the sum of the weights in each word identical to the lexical word is calculated as the similarity S. Specifically, the similarity S is calculated according to the following equation 2.

[Equation 2]
S (q, d) = g (ΣW (qi, dj), q, d)

  In addition, the determination means 352 can illustrate the following addition methods in order to calculate the similarity S of each word quickly and efficiently for many words. That is, a predetermined word In1 is specified among each word InN (N is a natural number). Then, among the words d1,..., Dt of the text information 341A1 of the predetermined word In1, the weights W (q1, In1) of the words d1,. Is added to the storage area s (In1) of the similarity S for each word provided in advance. Similarly, the processing described above is repeated for the lexical words q2 to qn, and processing for adding to the storage area s (In1) is performed. Further, similarly for the other words In2 to InN, a weighting factor W (qi, Ini) is sequentially calculated and added to the storage area s (In2) to the storage area s (InN) of the server storage unit 340. To do. Thus, the similarity S of each word is calculated and calculated as the identity of each document information 341A. The server storage unit 340 stores the identification result 341A2 of the document information 341A in association with the similarity S of the calculation result.

  The document selection unit 353 selects a candidate for the document information 341A having a content similar to the content of the lexical phrase based on the identity determined by the determination unit 352. Specifically, a plurality of pieces of identification information 341A2 having a number set in advance in descending order of the degree of similarity S calculated as identity by the determination unit 352 or a degree of similarity S equal to or greater than a predetermined threshold are selected. Note that the number of document information 341A to be selected may be set in advance in the server apparatus 300, or may be set as a condition for classification recognition processing by input setting in the terminal apparatus 200. The identification information 341A2 of the selected document information 341A is output to the classification acquisition unit 354.

  Based on the identification information 341A2 of the document information 341A selected by the document selection unit 353, the classification acquisition unit 354 reads the classification information 342A1 associated with the identification information 341A2 from the document-related classification storage area 342 of the server storage unit 340. Acquired as a candidate for classification for a document subject to processing request for classification recognition. Candidates for this classification are selected based on a score as a score corresponding to the similarity S calculated by the determination means 352, that is, a score calculated corresponding to the degree of similarity to the content of the lexical. This score calculation is a calculation of the appearance frequency for each classification in each classification information 342A1. For example, for each classification information 342A1 associated with the identification information 341A2, a score is calculated based on the similarity S associated with the identification information 341A2, and the score is calculated for each classification to calculate a total score. That is, the total score T is calculated by the following formula 3. In addition, as the calculation of the appearance frequency, any method can be used in addition to the calculation of the total score T according to this equation 3.

[Equation 3]
T = SUMk (f (simm (di, q)) × g (di, c))
k: Number of identification information 341A2 of the document information 341A selected by the document selection means 353
simm (di, q): Similarity between the document to be processed and the i-th document information 341A
f (s): Arbitrary function defined during operation that monotonically increases to the value of s
g (di, c): “0” when class c is not assigned to d,
“1” when touched as the first item,
A function that is a real number of “0 <g (d, c) <1” when used as the second item or less

  Here, the document information 341A is associated with the classification information 342A1 of the classification recognized by the execution of the classification recognition process, and is newly stored in the document information storage area 341 by the storage control unit 359 as described later. In this case, the total score T may be calculated to a value lower than the preset document information 341A, for example, 80%. Specifically, the total score T is calculated according to FIG. 4 below. Note that the newly stored document information 341A can be recognized based on flag information associated with the document information 341A, as will be described later in detail.

[Equation 4]
T = SUMk (f ′ (simm (di, q), New (di)) × g (di, c))
New (di): “1” when the document information 341A is newly registered,
If it is set in advance, set it to “0”.
f ′ (s, 0) = f (s)
It is assumed that f ′ (s, 1) = f (s) × 0.8. (Example set to 80%)

  Based on the calculated total score T, the classification acquisition unit 354 selects from the classification information 342A1 as a classification candidate for the document to be processed. That is, a plurality of pieces of classification information 342A1 having a preset number or a value equal to or greater than a predetermined threshold are selected in descending order of the total score T. The number of classification information 342A1 to be selected may be set in advance in the server apparatus 300, or may be set as a condition for classification recognition processing by input setting in the terminal apparatus 200. The selected classification information 342A1 is output to the notification unit 355.

  The notification unit 355 performs a process of displaying a list of the selected classification information 342A1 as a classification candidate on the terminal display unit 230. For example, the notification unit 355 creates a list display output form (not shown) in which the classification information 342A1 displayed as a list by the input operation in the terminal input unit 220 can be selected and modified. And the alerting | reporting means 355 performs the process which transmits the list display output form to the terminal device 200 via the network 110 by the server communication means 310, and displays it on the terminal display means 230. Further, the notifying unit 355 sends a setting input in which a predetermined synonym is selected or corrected by the input operation of the terminal input unit 220 according to the list display output form displayed on the terminal display unit 230 of the terminal device 200 via the network 110. Recognize. Based on the recognized setting input, the notifying unit 355 sets the classification information 342A1 selected or corrected by the input operation on the terminal device 200 as a classification for the classification recognition target document. A signal indicating that the classification information 342A1 has been set is output to the storage control means 359.

  The search unit 356 searches for the document information 341A from the document information storage area 341 of the server storage unit 340 and displays the document information 341A on the terminal display unit 230 by an input operation indicating that it is a document disclosure request in the terminal input unit 220. . The disclosure request of this document includes, for example, selection of classification information 342A1 displayed in a list by the notification means 355, in addition to input of incidental information and keyword search based on a predetermined input form. Then, the body information 341A1 of the retrieved document information 341A is transmitted to the terminal device 200 based on a predetermined output form by the notification unit 355, and displayed on the terminal display unit 230 so that it can be printed out by a connected printing device (not shown). Is done.

  The accounting calculation unit 357 calculates a consideration related to the notification of the result to the requesting user based on the request information and the document information 341A searched by the search unit 356 and transmitted by the notification unit 355. The calculation of the value is calculated based on, for example, the time when the terminal device 200 is connected to the server device 300 via the network 110, the processing fee for processing execution, the answer output fee corresponding to the amount of information of the result, and the like. . The amount of consideration is transmitted together with the result information such as the classification information 342A1 and the document information 341A so that the terminal device 200 can confirm the amount.

  The settlement unit 358 performs processing for collecting the consideration calculated by the billing calculation unit 357. For example, an invoice is issued based on customer information for identifying a user who uses a document search using the terminal device 200 stored in advance in the server storage unit 340, or each financial institution via the network 110 The server display means 330 indicates that payment processing is performed using farm banking (Firm Banking: FB) built in a so-called financial VAN which is a value-added network (VAN) managed by the server, or that payment processing is performed. To display an instruction for prompting the administrator to make a settlement process.

  When the storage control unit 359 recognizes the signal indicating that the classification information 342A1 is set from the notification unit 355, the storage control unit 359 acquires the document in which the classification information 342A1 is set from the terminal device 200 via the network 110, and the text information 341A1. And Then, the storage control unit 359 newly sets the identification information 341A2 and generates new document information 341A in association with the text information 341A1, and associates flag information indicating that the information is newly set and registered with the document information. The data is stored in the storage area 341. Further, the storage control unit 359 generates the document classification correspondence information 342A by associating the set classification information 342A1 with the new identification information 341A2, and stores the flag information in the document related classification storage area 342 in association with the flag information. Based on this flag information, the classification acquisition unit 354 can select the classification information 342A1 associated with the document information 341A stored in advance as compared with the newly stored document information 341A so that the classification acquisition unit 354 can select the classification information 342A1. Formula 4 is used as the calculation of the total score T.

[Operation of document information classification system]
Next, an operation of processing for setting a classification for a predetermined document in the above-described document information classification system 100 will be described with reference to the drawings. FIG. 6 is a flowchart showing the operation of search processing for searching for document information.

  When setting the document classification using the document information classification system 100, various information can be transmitted / received via the server device 300 and the network 110 by the user first appropriately inputting the terminal input unit 220 of the terminal device 200. The terminal device 200 is connected to the terminal. As this connection, for example, access to a homepage or the like in which the server device 300 that is built by installing application software for the server device in advance is resident can be exemplified. Then, the user appropriately inputs the terminal input unit 220 according to the screen display displayed on the terminal display unit 230, and inputs a lexical phrase included in the document for setting the document classification. By this input operation, the lexical setting unit 251 of the terminal processing unit 250 corresponds to the input operation from the terminal input unit 220, for example, the keyword appropriately converted corresponding to the signal sequence output for each operation of the keyboard operation key And a phrase such as a phrase are set (step S101).

  Thereafter, the user appropriately operates the terminal input unit 220 in accordance with the screen display, and executes the classification recognition process to the server apparatus 300 by an input operation for requesting a process for recognizing the classification of the document based on the set lexical phrase. Let Examples of the input operation for requesting execution of the classification recognition process include an input operation of a command button provided on the display screen and prompting the user to set the execution request for the classification recognition process. By this input operation, the request setting unit 252 of the terminal processing unit 250 generates request information for requesting the server apparatus 300 to execute processing for recognizing a classification by searching for documents similar to the set lexical phrase ( In step S102, the terminal communication unit 210 transmits the data to the server device 300 via the network 110 (step S103).

  When the server device 300 receives the request information transmitted from the terminal device 200 in step S103 (step S104), the request information acquisition unit 351 of the server processing unit 350 converts the lexical phrase included in the request information. Obtain (step S105). Based on the lexical acquired in step S105, the server processing unit 350 performs a similar document search step of searching for similar document information 341A for the lexical.

  That is, the determination unit 352 of the server processing unit 350 recognizes the lexical phrase acquired by the request information acquisition unit 351, and the text information 341A1 of the document information 341A stored in the document information storage area 341 of the server storage unit 340. Are identified (step S106). That is, the weighting degree information regarding the identity between the lexical phrase and the text information 341A1 of the document information 341A is calculated based on the above-described formula 1. Further, based on the weight information, the similarity S as a score related to the similarity state of the text information 341A1 with respect to the lexical is calculated based on the equation 2, and the similarity S is calculated based on the document to be classified and the text of the document information 341A. This is determined as the same content as the information 341A1. Based on the result of the identity calculation in step S106, the document selection unit 353 of the server processing unit 350 selects candidates for the document information 341A having contents similar to the contents of the lexical phrase, for example, the similarity S in a preset number. Are selected in descending order (step S107).

  After performing the similar document search process by the processing of step S106 and step S107, the server processing unit 350 performs a classification recognition process for recognizing the classification based on the searched document information 341A. In other words, based on the identification information 341A2 of the document information 341A selected in step S107, the classification acquisition unit 354 of the server processing unit 350 performs the classification associated with the identification information 341A2 from the document-related classification storage area 342 of the server storage unit 340. The information 342A1 is read in correspondence with the identity, and is acquired as a candidate for classification for the document for which classification recognition processing is requested (step S108). That is, for each classification information 342A1 associated with the identification information 341A2, a score based on the similarity S corresponding to the degree of similarity to the content of the lexical word is calculated, and the score is totaled for each classification using Expression 3 or Expression 4. Then, the total score T is calculated. Then, for example, the classification information 342A1 is selected as a candidate for classification in descending order of the total score T in a preset number.

  After the classification recognition step in step S108, the charging calculation unit 357 of the server processing unit 350, for example, the time when the terminal device 200 is connected to the server device 300 via the network 110, the fee for executing the classification recognition process, Based on the answer output fee corresponding to the information amount of the search result, the consideration regarding the notification of the search result is calculated for the requesting user (step S109). Thereafter, the server processing unit 350 performs a process of transmitting the classification information 342A1 selected by the notification unit 355 to the terminal device 200 so that the classification information is displayed on the terminal device 200 (step S110). The terminal device 200 to which the selected classification information 342A1 is transmitted is specified based on the request information received in step S104, for example.

  In step S110, when the terminal device 200 receives the list display form related to the classification information 342A1 selected by the server device 300 by the terminal communication unit 210 (step S111), the output control unit 254 of the terminal processing unit 250 uses the terminal display unit. 230 is appropriately controlled, and the classification information 342A1 is displayed on the screen so that it can be selected and modified by an input operation with the terminal input means 220 (step S112). When the predetermined classification information 342A1 is selected or corrected according to the screen display of the list display form (step S113), the output control means 254 causes the server apparatus 300 to transmit the classification information 342A1 selected or corrected by the input operation. (Step S114), the document classification recognition process in the terminal device 200 ends.

  Further, when the classification information 342A1 selected or modified in step S114 is received by the server communication unit 310 from the terminal device 200 (step S115), the notification unit 355 of the server processing unit 350 selects or modified the classification information. A process of determining 342A1 as the classification of the document to be recognized and storing the new document information 341A in the server storage unit 340 is performed (step S116). That is, when the notification unit 355 recognizes that the classification information 342A1 is set, the storage control unit 359 acquires the document in which the classification information 342A1 is set from the terminal device 200 via the network 110 and generates the text information 341A1. . Further, the storage control unit 359 newly sets the identification information 341A2 and generates new document information 341A in association with the text information 341A1, and associates flag information indicating that the information is newly registered and registered with the document. The information is stored in the information storage area 341. Further, the storage control unit 359 generates the document classification correspondence information 342A by associating the set classification information 342A1 with the new identification information 341A2, and stores the flag information in the document related classification storage area 342 in association with the flag information.

  Thereafter, the server apparatus 300 performs a process for collecting the consideration calculated by the charging calculation means 357 by the settlement means 358 (step S117). For example, an invoice is issued based on customer information for identifying a user who uses a document search using the terminal device 200 stored in advance in the server storage unit 340, or so-called financial VAN via the network 110. The server apparatus 300 performs processing such as payment processing using the farm banking constructed in the above, or displays the fact that payment processing is to be performed on the server display means 330 and notifies the administrator of guidance for payment processing. The classification recognition process ends.

[Effects of Embodiment]
As described above, in the above embodiment, when the request information acquisition unit 351 acquires lexical information related to a lexical word included in a document to be recognized for classification, the classification information 342A1 related to the classification is associated with each other and constructed in one data structure. The determination unit 352 determines the identity between the lexical word included in the document information 341A stored in the document information storage area 341 having a table structure storing a plurality of the document information 341A and the acquired lexical word. Based on the determined identity, the document information 341A stored in the document information storage area 341 is selected by the document selection means 353, and the classification information 342A1 associated with the selected document information 341A is recognized by the classification acquisition means 354. Get as target classification. As described above, a document having a content similar to that of the document to be recognized for classification is recognized by judging the identity of the lexical characters included in the document based on the document information 341A associated with the classification information 342A1 in advance, and the content is similar. The document classification is recognized as the classification of the document to be recognized. For this reason, for example, using the database of the document information 341A in which the existing classification is recognized, it is possible to easily recognize the classification of the documents by an easy calculation.

  Then, the request information acquisition unit 351 acquires a plurality of words as lexical information, and of the acquired plurality of words, the ratio of the same words as the words included in the document information 341A in the document information storage area 341, for example, lexical information The similarity S is calculated by the determination means 352. For this reason, for example, it is possible to easily recognize the similarity of the contents of the document by determining the proportion of the recognized words by morphological analysis as the lexical identity, and to easily determine the similarity of the contents of the document by an easy calculation. Can recognize the classification.

  Further, the score information related to the score corresponding to the identity determined by the determining means 352 is associated with the document information 341A in the document information storage area 341, and the document information 341A associated with the high score information is selected by the document selecting means 353. ing. Therefore, the degree of similarity of the document information 341A whose contents are similar to the document to be recognized is easily recognized by the score, and the identification of the document information 341A whose contents are similar can be obtained by a simple calculation, and the document information 341A whose contents are similar. The appropriate classification can be easily recognized based on the above.

  Then, the classification acquisition unit 354 calculates the appearance frequency for each classification of the classification information 342A1 associated with the document information 341A selected by the document selection unit 353, and recognizes the classification information 342A1 of the classification with the high appearance frequency of each classification. It is classified as a target. For this reason, the suitable classification corresponding to the contents can be recognized. That is, since the classification acquisition unit 354 counts the classification information 342A1 in each document information 341A selected by the document selection unit 353 for each classification and calculates the appearance frequency, a suitable classification corresponding to the content is simpler. It is easily recognized by simple calculation. Specifically, the classification acquisition unit 354 acquires the classification information 342A1 and the score information in the document information 341A selected by the document selection unit 353. For example, as shown in Equation 3 or Equation 4, the classification of the acquired classification information 342A1 is obtained. The total score T is calculated by adding the scores of the score information every time, and the classification information 342A1 having a high total score T is acquired as the classification to be recognized. For this reason, it is possible to easily select an appropriate classification according to the content similar situation by a simple calculation method.

  In addition, the notification unit 355 of the server device 300 causes the terminal display unit 230 to display the classification acquired by the classification acquisition unit 354 so that the classification can be selected by an input operation. For this reason, the user can confirm the classification recognized by the calculation, and can be appropriately set and corrected, thereby obtaining a more appropriate classification setting.

  The storage control unit 359 associates the classification recognized by the server processing unit 350 with the document information 341A to be recognized, and stores this document information 341A in the document information storage area 341. For this reason, the document information 341A is automatically accumulated, the usability can be improved, the document information storage area 341 can be easily updated, and system management can be facilitated. Furthermore, for example, by constructing the document information 341A for each classification, processing such as searching the document information 341A from the classification becomes easy, and the construction of the document information storage area 341 that is easy to use can be facilitated.

  Further, the document information 341A selected by the document selection unit 353 is associated with identification information indicating that the storage control unit 359 associates the classification with the document information 341A to be stored in the document information storage area 341 and is newly stored. For this reason, as shown in Equation 4, for example, the processing for preferentially acquiring the classification information 342A1 of the document information 341A not associated with the identification information by the classification acquisition unit 354 can be easily performed, and the recognized classification is associated with the new information. Thus, the recognition result of the classification is less likely to be generated by storing the document information as the document information 341A for recognizing the classification and constructing the document information storage area 341, so that the classification can be recognized with higher reliability.

  As the classification of the classification information 342A1, for example, IPC, which is an international patent classification that is a classification of contents that are internationally unified, is used. For this reason, a uniform classification can be obtained, a database such as patent publication gazette data that is an existing patent document associated with an international patent classification can be used, and it is not necessary to construct a document information storage area 341 in advance, which is easy. The classification can be easily recognized with a simple configuration.

In addition, the system configuration is such that the server device 300 performs processing for recognizing the classification for the requested document by the server device 300 in response to a classification recognition processing request from the terminal device 200 connected via the network 110 and notifies the terminal device 200 of the processing. For this reason, even with a simple configuration of the terminal device 200, it is possible to easily recognize an appropriate classification, and it is possible to easily expand use. Furthermore, the server apparatus 300 can manage the document information 341A having a large capacity and various tables in an integrated manner, and data can be easily updated, so that the operational management can be easily improved. Therefore, the processing capacity is relatively small storage capacity, such as mobile phones also be used as a relatively small it is possible to easily recognize the classification of documents, thereby easily be expanded for use.

  The personal computer or the like is configured to be able to execute classification recognition processing for documents as a program. For this reason, by using a recording medium in which the program is incorporated or the program is recorded, the above processing can be easily obtained, and the use can be easily expanded.

[Modification of Embodiment]
Note that the present invention is not limited to the above-described embodiment, and includes the following modifications as long as the object of the present invention can be achieved.

  For example, although described as a system configuration, a terminal is used by determining the lexical identity with a recording medium having a document information storage area 341, for example, document information 341A stored in the terminal storage unit 240, using a personal computer. The display unit 230 displays the classification, the document information storage area 341 is read via the network 110, and the terminal processing unit 250 determines the identity. The server device 300 determines the identity. A system configuration in which the document information 341A is selected and a classification is selected by the terminal device 200 based on the document information 341A, or the server storage unit 340 is separated from the server device 300 and is separated from the server device 300 through another network 110. It can be in any form such as being configured as a server device.

  Further, as the determination of the identity, for example, the similarity S may be calculated from a sequence of character strings obtained by shifting the character string of the lexical information by one character at a time. That is, for example, a three-character partial character string shifted by one character, a two-character partial character string, or a single character unit character string is generated, and the partial character string or unit character string and the text information of the document information 341A are generated. The weighting degree information on the identity with 341A1 is calculated based on Formula 1, and the similarity S as a score regarding the similarity state of the text information 341A1 with respect to the lexical is calculated based on Formula 2 based on the weighting information, The identity of the lexical information and the document information 341A may be determined by any method, such as identity. Note that the configuration in which the identity is determined by the arrangement of character strings shifted by one character reduces search omissions, and the identity of lexical characters can be appropriately determined. In addition, it is not necessary to perform analysis such as morphological analysis, and it is possible to easily determine appropriate lexical identity with simple processing. The number of characters in the character string is not particularly limited. For example, it may be a three-character partial character string shifted by one character, a two-character partial character string, or a single character unit character string. Furthermore, these three-character partial character string, two-character partial character string, and It may be configured to extract all unit character strings of one character.

  Then, when the classification information 342A1 displayed on the terminal display unit 230 is selected by an input operation, an explanatory text regarding the contents of the classification may be displayed. Specifically, a classification information storage area having a table structure in which a plurality of pieces of explanatory information such as explanatory texts related to classification contents are associated with the classification information 342A1 in the server storage unit 340, and a plurality of pieces of classification information storage area are stored. When transmitting to the terminal device 200, the description information is also transmitted, or when the server device 300 that has recognized the selected input operation via the network transmits the description information to the terminal device 200, the output control means 245 To display on the terminal display means 230. With this configuration, the content of the recognized classification can be easily confirmed by the user, and usability can be improved.

  Further, the classification displayed on the terminal display unit 230 may be displayed in a display form corresponding to the identity. Specifically, when transmitting the classification information 342A1 to the terminal device 200 by the notification unit 355, the information of the total score T for each classification based on the similarity S, which is information related to the identity, is also transmitted and output control is performed. A means 245 performs display processing corresponding to the total score T on the terminal display means 230, for example, color and brightness, etc., as appropriate, and display in a display form in which the visibility increases as the value of the total score T increases. . With this configuration, the user can easily recognize the conformity state with the content of the document information 341A to be recognized, which is particularly effective when, for example, a plurality of classifications are displayed and notified.

Furthermore, the classification displayed on the terminal display unit 230 may be displayed together with the score corresponding to the identity. Specifically, as in the case of displaying in the display form corresponding to the above-described identity, for example, the information unit 355 causes the information of the total score T to be transmitted to the terminal device 200 together with the value of the total score T. The output control means 245 performs a process of displaying the classification on the terminal display means 230 in a state where the classification is written together. This configuration allows the user to easily recognize the matching state with the content of the document information 341A to be recognized, similarly to the configuration of different display forms, and is particularly effective when, for example , a plurality of classifications are displayed and notified. It is.

  Further, the literary information of the lexical information acquired by the request information acquisition unit 351 may be displayed in a display form different from other lexical information. Specifically, the lexical phrase set by the lexical setting means 251 is stored in the terminal storage means 240, and when the document to be recognized for classification is displayed on the terminal display means 230, the set lexical phrase is another lexical phrase. Different display forms, such as color and brightness, are set as appropriate so that the user can easily recognize them. Further, the lexical phrase acquired by the request information acquisition unit 351 is stored in the server storage unit 340, and predetermined document information 341A is searched and notified to the terminal device 200 in response to a request such as document search from the terminal device 200. At this time, the document information 341 </ b> A may be transmitted to the terminal device 200 by the notification unit 355 in a state where the stored lexical phrase is in a different display form as described above. With such a configuration, for example, it is possible to make the user easily recognize lexical characters that cause classification recognition in the document information 341A to be recognized for classification and the document information 341A in the same classification as the document information 341A. Can be improved.

  Further, in the configuration in which the document information 341A is displayed so that the lexical words in the classification recognition process have different display forms, the classification of the classification information 342A1 is performed by an input operation for selecting the lexical words in the different display forms. For example, a classification name such as IPC or an explanation may be displayed. With such a configuration, it is possible to easily confirm the classification of the document information 341A and improve usability.

  Further, the literature is not limited to publications in industrial property rights, but can also be various publications such as newspapers, magazines, and various books. Furthermore, it is not limited to the configuration extracted by morphological analysis, but may be a lexical phrase set and input in advance by key operation.

  Furthermore, as the identity, the value obtained by calculating the similarity S based on the weight W and scoring is described as the identity, but is not limited to this method. Moreover, it is not necessary to score.

  In addition, the document whose classification is recognized is described in the document information storage area 341, but it may not be stored. Further, in the case where the document is stored, the document stored in advance is selected preferentially over the newly stored document. However, the document may be selected equally. Furthermore, although the description has been made in association with the identification information indicating that the information is newly stored as a configuration having selection priority, the present invention is not limited to this case, and the newly stored information can be distinguished from the previously stored information. Either configuration can be used.

  In addition, the selected classification is displayed on the terminal device 200 for notification, but the configuration for notification is not limited to display, and any configuration for notification by voice, printed matter, or the like can be used.

  In addition, the present invention is not limited to the above-described embodiment and modifications of the embodiment, and other modifications such as changing the data structure or setting data structure items are possible without departing from the object of the present invention. Various applications are possible, such as changing the configuration for processing.

  INDUSTRIAL APPLICABILITY The present invention can be used for a configuration in which a classification such as an international patent classification for document information related to industrial property rights in a broad sense such as a trade name, as well as patents, utility models, designs, trademarks, and the like is arbitrarily described. Further, as the literature information, publications other than the gazette can be targeted. Furthermore, the classification is not limited to the international patent classification. For example, the contents of bibliographic information such as “report”, “sports”, “entertainment”, “medicine”, “mathematics”, “biology”, etc. Any classification as specified can be targeted.

It is a conceptual diagram which shows schematic structure of the literature information classification system which concerns on one embodiment of this invention. It is a block diagram which shows schematic structure of the terminal device in the said embodiment. It is a block diagram which shows schematic structure of the server apparatus in the said one Embodiment. It is a schematic diagram which shows notionally the table structure of the literature information storage area in the said one Embodiment. It is a schematic diagram which shows notionally the table structure of the literature relevant classification memory area in the said one Embodiment. It is a flowchart which shows the operation | movement of the process which recognizes the classification | category of the literature in the said one Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Literature information classification system as literature classification device 230 ... Terminal display means as display means 254 ... Output control means which also functions as display control means
340 ... server storage section 341 ... literature information storage area 341A ... document information as document storage means
342 ... literature related classification storage area 350 ... request information acquisition unit 352 ... determining unit 353 ... Document selecting means 354 ... classification acquisition unit that also functions as a server processing unit 351 ... acquisition means constituting the document classification recognizer as an arithmetic means 355 ... Notification means that also functions as display control means 359 ... Storage control means

Claims (19)

Words a recognized document classification recognizer classification including plain coarse Document information,
A resulting unit preparative for obtaining a word included in the recognition the document information of the target of the classification,
And words included in the document information and the the classification and configured document information stored in the document storage means stores a plurality associated comprise a plurality of words, the word acquired by the acquisition means A determination means for determining identity;
Literature selection means for selecting the literature information stored in the literature storage means based on the identity;
A class acquisition means for calculating a score based on the determined identity from the class associated with the selected document information and selecting a candidate for the class based on the total score for each class; Equipped,
The determination means includes
For a word in the document information stored in the document storage unit that matches the acquired word, the matching word appears in the document information stored in the document storage unit and the document information to be recognized. and the number and match the number of words, the total number of words constituting said literature document information of document information stored in the storage means and the recognition target and the total number of words, dividing the total number of words in the word-match count By calculating the weight information as the identity,
The classification acquisition means includes
The document classification recognition apparatus, wherein the score is calculated based on weight degree information calculated for the word. The document classification recognition apparatus according to claim 1,
The document classification means characterized in that, when there is flag information indicating that the information is newly set and registered associated with the document information, the document acquisition unit performs a calculation to reduce the total score for each of the categories Recognition device. The document classification recognition apparatus according to claim 1 or 2 ,
The determination means associates score information regarding the score corresponding to the identity with the document information of the document storage means,
The document classification recognition apparatus, wherein the document selection means selects the document information having a high score of the score information. The document classification recognition apparatus according to claim 3 ,
The classification acquisition unit, the score of the both the classification associated with the document information that has been selected in the literature selecting means acquires in association with the score information, the score information to the each classification acquired from the document information A document classification recognition apparatus characterized in that a total score is calculated by adding together and the classification having a high total score is acquired as the classification of the recognition target. The document classification recognition apparatus according to claim 1 or 2 ,
The classification acquisition unit, the calculates the appearance frequency of obtained each classification associated with the document information that has been selected in the literature selecting means, obtaining a frequency of occurrence of each of these classifications higher classification as the classification of the recognition target A document classification recognition device characterized by the above. The document classification recognition apparatus according to claim 5 ,
The said classification acquisition means counts the classification | category in each literature information selected by the said literature selection means for every classification , and calculates appearance frequency. The literature classification recognition apparatus characterized by the above-mentioned. A document classification recognition apparatus according to any one of claims 1 to 6 ,
The document classification recognition apparatus according to claim 1 , wherein the classification is an international patent classification. A document classification recognition apparatus according to any one of claims 1 to 7 ,
A document classification recognition apparatus, comprising: a display control unit configured to display the classification acquired by the classification acquisition unit on a display unit so as to be selectable by an input operation. The document classification recognition apparatus according to claim 8 ,
When the display control means recognizes that the classification to be displayed on the display means has been selected by an input operation, the display control means causes the display means to display explanatory information relating to the contents of the classification. The document classification recognition apparatus according to claim 8 or 9 , wherein
The document classification recognition apparatus characterized in that the display control means controls to display the classification acquired by the classification acquisition means in a display form corresponding to the identity. A document classification recognition apparatus according to any one of claims 8 to 10 ,
The document classification recognition apparatus, wherein the display control means controls to display the classification acquired by the classification acquisition means together with the score information related to the score corresponding to the identity. A document classification recognition apparatus according to any one of claims 1 to 11 ,
When displaying on the display means a word acquired by the acquisition means is characterized by comprising a display control means for displaying on the display means the document information in the display mode different from the case of displaying the other words Document classification recognition device. The document classification recognition apparatus according to claim 12 ,
Wherein the display control unit, document classification recognition apparatus, characterized in that on the display unit was acquired classified the words displayed on the different display form and the other word is selected by an input operation by the word. A document classification recognition apparatus according to any one of claims 1 to 13 ,
Data distribution operation including at least one of changing the document information transmitted from a server device connected via a network and stored in the document storage unit and changing the storage state of the document information Recognizing operation information related to the data storage means, the storage means for controlling the reference storage means to control the change of the reference information and the change of the storage status in response to the data distribution operation. Document classification recognition device. And document information includes a plurality of words, the document storage means classification and are more stored in association with,
The document classification recognition apparatus according to any one of claims 1 to 14 , wherein the document storage means is used to recognize a classification of predetermined document information.
A document classification apparatus characterized by comprising: The document classification device according to claim 15 ,
A document classification apparatus comprising storage control means for associating a classification recognized by the document classification recognition apparatus with the document information to be recognized and storing the document information in the document storage means. A computer, a word a sort recognizes Document classification recognition method including plain coarse Document information,
The computer
When acquiring the words included in the document information of the recognition target of the classification, the document information stored in the document storage means stores a plurality associated with it and the classification to be composed document information including a plurality of words The identity of the word contained in and the acquired word ,
When the identity is determined, for the word in the document information stored in the document storage unit that matches the acquired word , the document information in which the matching word is stored in the document storage unit and the recognition target of the word-match count the number of occurrences in the document information, the total number of words constituting said literature document information of document information stored in the storage means and the recognition target and the total number of words, the total number of words Is divided by the number of matching words to calculate the weight information as the identity,
Based on the determined identity, the document information stored in the document storage means is selected,
From the classification associated with the selected document information, a score is calculated based on the weight information as the calculated identity, and the classification candidate is selected based on the total score for each classification. Document classification recognition method characterized by A document classification program that causes a computer to function as the document classification recognition device according to any one of claims 1 to 14 or the document classification device according to claim 15 or claim 16 . A literature classification program for causing a computer to execute the literature classification recognition method according to claim 17 .

Images