JP3771047B2 - Document classification apparatus, document classification method, and computer-readable recording medium storing a program for causing a computer to execute the method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a document classification device that classifies documents based on the contents of the document, a document classification method, and a computer-readable recording medium that records a program that causes a computer to execute the method.
[0002]
[Prior art]
Conventionally, as a document classification device, for example, in the document classification device described in JP-A-7-36897, a document is regarded as a document vector characterized by a word, and these document vectors are grouped using a clustering method, It describes what performs automatic classification of documents.
[0003]
In addition, document data is generally stored in a database, and not only the document content but also bibliographic items such as creation date and creator are added, or the document content itself contains multiple items There are many. For example, a patent publication is composed of a plurality of items such as “claims” and “detailed description of the invention”.
[0004]
[Problems to be solved by the invention]
However, since the above-described prior art document classification apparatus cannot arbitrarily specify an item to be classified by an operator for document data having a plurality of items, data that adversely affects classification is added. In addition, since a plurality of items cannot be combined, there is a problem that data effective for classification is insufficient and a highly accurate classification result cannot be obtained.
[0005]
The present invention eliminates the problems caused by the above-described conventional example, and provides a document classification device, a document classification method, and a program for causing a computer to execute the method that can perform highly accurate classification reflecting the intentions of an operator. An object is to provide a recorded computer-readable recording medium.
[0006]
[Means for Solving the Problems]
In order to solve the above-mentioned problems and achieve the purpose, this The document classification device according to the invention is a document classification device that performs document classification based on the content of a document, an input unit that inputs document data composed of one or more items, and an input unit that inputs the document data A designation unit for designating the items constituting the document data, a conversion unit for converting the document data so that only the data corresponding to the item specified by the specification unit is included, and the conversion unit Classification means for classifying documents using the converted data.
[0007]
This Departure According to Meiji, since only the content data of the designated item is used when classifying the document, it is possible to prevent the content of other items from affecting the classification result. For this reason, it is possible to efficiently perform classification with high accuracy close to the classification desired by the operator by designating the items of document data that are considered necessary from the viewpoint of the classification expected by the operator.
[0008]
Also, This invention The document classification apparatus according to claim 1, wherein the document classification apparatus classifies the document based on the content of the document, the input means for inputting the document data composed of one or a plurality of items, and the document input by the input means Designation means for designating the items constituting the data, conversion means for converting the document data so that only the data corresponding to the item designated by the designation means is included, and the conversion converted by the conversion means Document vector generation means for generating feature vectors of each document using data, and classification means for classifying documents using the feature vectors of each document generated by the document vector generation means, To do.
[0009]
This Departure According to Meiji, only the content data of the specified item is used when generating the feature vector of each document for classifying the document, so that the effect of the content of other items on the classification result can be prevented. it can. For this reason, it is possible to perform classification with high accuracy close to the classification desired by the operator by designating the items of the document data that are considered necessary from the viewpoint of the classification expected by the operator.
[0010]
Also, This invention The document classification device according to the above The invention is characterized in that, when the conversion means converts the document data, a predetermined symbol is inserted between the data of the items so that the data of the items can be separated.
[0011]
This Departure According to Akira, since a delimiter symbol is inserted between each conversion data, when performing analysis processing such as morphological analysis, the conversion data as a whole is completely different by combining the data corresponding to each item as they are. It is possible to avoid the meaning being constructed.
[0012]
Also, This invention According to the document classification method according to the document classification method for classifying documents based on the contents of the document, an input step of inputting document data composed of one or a plurality of items, and a document input by the input step A designation step for designating the items constituting the data, a conversion step for converting the document data so that only the data corresponding to the items designated by the designation step is included, and the conversion converted by the conversion step And a classification step of classifying documents using data.
[0013]
This Departure According to Meiji, since only the content data of the designated item is used when classifying the document, it is possible to prevent the content of other items from affecting the classification result. For this reason, it is possible to perform classification with high accuracy close to the classification desired by the operator by designating items of document data that are considered necessary for the viewpoint of the classification expected by the operator.
[0014]
Also, This invention According to the document classification method according to the document classification method for classifying documents based on the contents of the document, an input step of inputting document data composed of one or a plurality of items, and a document input by the input step A designation step for designating the items constituting the data, a conversion step for converting the document data so that only the data corresponding to the items designated by the designation step is included, and the conversion converted by the conversion step A document vector generation step of generating feature vectors of each document using data, and a classification step of classifying documents using the feature vectors of each document generated by the document vector generation step, To do.
[0015]
This Departure According to Meiji, only the content data of the specified item is used when generating the feature vector of each document for classifying the document, so that the effect of the content of other items on the classification result can be prevented. it can. For this reason, it is possible to perform classification with high accuracy close to the classification desired by the operator by designating items of document data that are considered necessary for the viewpoint of the classification expected by the operator.
[0016]
Also, This invention The document classification method related to the above In the invention, the conversion step inserts a predetermined symbol between the data of the items so that the data of the items can be separated when converting the document data.
[0017]
This Departure According to Ming, since a delimiter symbol is inserted between each conversion data, it is possible to avoid confusing and handling data of multiple items as one data during analysis processing such as morphological analysis. The content data for each item can be instantly identified.
[0018]
Also, This The storage medium according to the invention is the above By recording a program that causes a computer to execute the method described in the above, the program can be machine-readable, the above Can be realized by a computer.
[0019]
DETAILED DESCRIPTION OF THE INVENTION
Exemplary embodiments of a document classification device, a document classification method, and a computer-readable recording medium recording a program that causes a computer to execute the method will be described below in detail with reference to the accompanying drawings.
[0020]
(Embodiment 1)
First, the hardware configuration of the entire information processing system constituting the document classification apparatus according to Embodiment 1 of the present invention will be described. FIG. 1 is an explanatory diagram showing the hardware configuration of the entire information processing system constituting the document classification apparatus according to the first embodiment. In FIG. 1, the information processing system constituting the document classification apparatus according to the first embodiment is configured by a server / client system. That is, the server 101 and a plurality of clients 102 are connected by the network 103.
[0021]
The client 102 generates classification data, instructs the server 101, displays the classification result, and the like. On the other hand, in accordance with an instruction from the client 102, the server 101 performs processing relating to document (text) classification by enormous numerical operations, and sends the processing result to the client 102. More specifically, the server 101 performs text classification processing, and the client 102 performs classification data generation, processing execution instruction, text classification result display, and the like.
[0022]
Data exchange between the server 101 and the client 102 uses a method called file sharing. That is, by creating a file used for classification processing in a shared folder on the server 101, the two exchange data. Therefore, the client 102 can use the shared folder of the server 101 by sharing the network.
[0023]
Next, the hardware configuration of the server 101 and the client 102 will be described. FIG. 2 is an explanatory diagram illustrating the hardware of the server 101 in the information processing system constituting the document classification device according to the first embodiment. As the server 101, for example, a workstation (WS) is used.
[0024]
In FIG. 2, 201 is a CPU that controls the entire server 101, 202 is a ROM that stores a boot program and the like, 203 is a RAM 203 that is used as a work area for the CPU 201, and 204 is a network 103 via a communication line 205. Reference numeral 206 denotes an interface (I / F) connected to the network 103 and serving as an internal interface, and a disk device 206 stores data. Reference numeral 200 denotes a bus for connecting the above-described units.
[0025]
In addition, a display 208 that displays document information, image information, function information, and the like, a keyboard 209 and a mouse 210 for inputting data, and the like may be similarly connected. Further, the disk device 206 is provided with a shared folder 207 for exchanging data with the client 102.
[0026]
FIG. 3 is an explanatory diagram showing the client 102 in hardware in the information processing system constituting the document classification device according to the first embodiment. As the client 102, for example, a personal computer (PC) or the like is used.
[0027]
In FIG. 3, 301 is a CPU that controls the entire system, 302 is a ROM that stores a boot program, 303 is a RAM that is used as a work area of the CPU 301, and 304 is an HD (hard disk) 305 according to the control of the CPU 301. HDD (Hard Disk Drive) for controlling the reading / writing of data with respect to the HDD, 305 for storing the data written under the control of the HDD 304, and 306 for reading / writing the data with respect to the FD (floppy disk) 307 according to the control of the CPU 301. FDD (floppy disk drive) for controlling writing, 307 for a removable FD for storing data written under the control of the FDD 306, and 308 for a display for displaying documents, images, function information, etc. That.
[0028]
Further, reference numeral 309 is connected to the network 103 via the communication line 310, and an interface (I / F) that controls an internal interface with the network 103, and 311 has keys for inputting characters, numerical values, various instructions, and the like. 312 has a mouse for moving the cursor, selecting a range, pressing an icon or button displayed on the display screen, moving a window, changing the size, etc., and 313 has an OCR (Optical Character Reader) function. 314 indicates a scanner for optically reading the image, 314 indicates a printer for printing the contents of data including the classification result, and 315 indicates a bus for connecting the above-described units.
[0029]
Next, a functional configuration of the document classification device according to Embodiment 1 will be described. FIG. 4 is a block diagram functionally showing the configuration of the document classification apparatus according to the first embodiment. In FIG. 4, the document classification apparatus includes an input unit 401, a specification unit 402, a conversion unit 403, a conversion data storage unit 404, a classification unit 405, and a classification result storage unit 406.
[0030]
Next, the contents of each component will be described in detail. The input unit 401, the specifying unit 402, the conversion unit 403, the conversion data storage unit 404, the classification unit 405, and the classification result storage unit 406 are a recording medium such as the ROM 202 or 302, the RAM 203 or 303, or the disk device 306 or the hard disk 316. The function of each unit is realized by the CPU 201 or 301 executing instruction processing according to the instructions described in the program recorded in the above.
[0031]
(Input unit 401)
The input unit 401 is for inputting document data. For example, the I / F 204 or 309 can obtain a document or a document group via the keyboard 209 or 311, the scanner 313 having the OCR function, or the network 103. Etc. In addition to the above, the input unit 401 includes all of them if it can acquire document data.
[0032]
For example, when the document data is stored in a database, the document data is also input when the medium in which the database is recorded is incorporated in the document classification apparatus according to the present embodiment. Furthermore, a document data storage unit (not shown) that stores the input document data may be included. The document data storage unit may be, for example, the disk device 206 of the server 101 having a large capacity memory.
[0033]
Here, in the present embodiment, a document is a collection of one or more sentences written in a natural language, and when it is a classification target, this is called a document. Specifically, an open patent gazette or a specific newspaper article is also a document, and a claim or a specific sentence is taken out as a document.
[0034]
(Designation unit 402)
The designation unit 402 designates an item of document data. Specifically, the designation unit is composed of three processes.
[0035]
First, an item is extracted from the document data input by the input unit 401 (first process). As a method of extracting an item, there is a method of searching for an item to which a predetermined code (for example, “[” “]” or the like) is added in advance and selecting the item.
[0036]
The first process may be performed by the input unit 401 instead of the designation unit 402. That is, when the input unit 401 inputs document data, the document data items are also extracted. The extraction result is stored in the document data storage unit in association with the document data. In this case, the first process is omitted in the designation unit 402 by using the extraction result. Some types of databases have information about items in advance, and the first processing is omitted by using information about the items.
[0037]
Next, based on the extraction result of the item by the first process, the extraction result stored in the document data storage unit, the information related to the item, etc., what is the extracted item, and the item A list of the contents corresponding to is presented to the operator (second process). As a method of presentation, there are a method of displaying only an item on the display 208 or 308, or displaying all or part of an item and contents corresponding to the item.
[0038]
As a method for displaying only the items, for example, there is a method in which the item names are displayed so as to be arranged in a vertical line in horizontal writing based on the appearance order in the document data. In this case, if the number of items is larger than the number of display lines on the display screen, it may be folded and displayed in two or more vertical columns, or displayed in one vertical column and the display screen can be scrolled vertically. You may do it.
[0039]
As a method of displaying all or part of the item and the content corresponding to the item, for example, the item name is displayed in horizontal writing based on the appearance order in the document data, as in the method of representing only the item. There is a method of displaying in a line so that the contents are displayed in a line, and the contents are also displayed in horizontal writing at the position arranged corresponding to the item name on the right side. In this case, when the data amount of the content is larger than the number of display columns on the display screen, the display screen may be scrollable in the horizontal direction.
[0040]
Another method for displaying all or part of an item and the contents corresponding to the item is to display only the item name, move the cursor to the area where the item name is displayed, and perform a predetermined operation (mouse All or part of the content data may be displayed in a pop-up manner by pressing a button 210 or 312 or a predetermined key such as the keyboard 209 or 311).
[0041]
Next, according to an instruction from the operator, one or more items to be classified are designated simultaneously from the presented (displayed) items (third process). As a designation method, a corresponding item is designated from the presented items based on an instruction signal related to designation from a pointing device such as the keyboard 209 or 311 or the mouse 210 or 312.
[0042]
At this time, one item may be specified, or two or more items may be specified at the same time. Also, the form of coupling can be specified together. Furthermore, the arrangement order of the content data after the data conversion may be designated according to the designation order.
[0043]
(Conversion unit 403)
The conversion unit 403 converts the document data so that the input document data includes only the data corresponding to the item specified by the specification unit 402. Specifically, only the data corresponding to the specified item in the document data is extracted and converted into converted data consisting only of the extracted data.
[0044]
The conversion data is not only converted by listing the data corresponding to each item in the order of the specified items in the original document data, but for example, the data contents of the specified items are combined as a character string Thus, conversion data including only the data content of the specified item may be used, or conversion may be performed such that the data is combined after the order of the items is changed to an order different from the order in the original document data.
[0045]
Also, the conversion unit 403 inserts a predetermined separation symbol 601 between the item data so that the data of each item in the conversion data can be separated. Thereby, the break of the data corresponding to each item can be grasped instantly.
[0046]
The separation symbol 601 is particularly important when performing natural language analysis such as morphological analysis. If the data corresponding to each item is in the form of a sentence (when the end of the sentence ends with a punctuation mark), it is possible to determine the break between the sentence and sentence without this separator. If the data corresponding to the item does not form the body of the sentence (in the case of a bulleted sentence, the item changes in the middle of the sentence, etc.), combining the data as they are has a completely different meaning depending on the item. May be configured. In order to avoid such a case, this separator 601 is inserted.
[0047]
In general, the separator 601 uses “/ (slash)” indicating a break, but if “/” is present in the converted data, confusion with “/” of the data occurs. You can use another symbol. Whether or not to insert this symbol, a predetermined key is assigned to the keyboard 209 or 311 and a predetermined icon is displayed each time the key is pressed or on the display screen. It is also possible to alternately set whether or not to insert the separation symbol 601 each time is clicked.
[0048]
(Conversion data storage unit 404)
The conversion data storage unit 404 is a storage unit that stores conversion data. The conversion data storage unit 404 can be set, for example, according to a difference in conversion data capacity or use, such as the disk device 206 of the server 101, the client-side hard disk 305, or the floppy disk 307.
[0049]
The conversion data storage unit 404 stores the separation symbol 601 and the like in addition to the conversion data including the item setting order. The conversion data stored in the conversion data storage unit 404 can be utilized, for example, when used in another classification.
[0050]
(Classification unit 405)
The classification unit 405 automatically classifies according to the content of the conversion data converted by the conversion unit 403 or the conversion data stored in the conversion data storage unit 404. The classification unit 405 can classify a document using a conventional document classification method such as “document classification device” disclosed in Japanese Patent Application Laid-Open No. 7-36897.
[0051]
(Classification result storage unit 406)
The classification result storage unit 406 is a storage unit that stores the results classified by the classification unit 405. Similar to the conversion data storage unit 404, the classification result storage unit 406 includes, for example, a disk device 206 of the server 101, a client-side hard disk 305, a floppy disk 307, etc. Each can be set.
[0052]
Next, an example of converted data obtained by converting document data and document data will be described. FIG. 5 is an explanatory diagram showing an example of document data and its conversion data. In FIG. 5, a patent publication group is used as a document group, 501 is an example of document data, and 502 is an example of converted data.
[0053]
The document data 501 includes “application number”, “application date”, “inventor”, “invention name”, “purpose”, “configuration”, “claim 1”, “prior art”, and “solve the problem”. Items such as “means for achieving”, “action”, “example”, “effect of the invention” are included.
[0054]
Since conventional document classification devices handle each piece of document data as a group, the content data of all items is subject to classification processing for document data containing multiple items, which is unnecessary for the classification viewpoint desired by the operator. Or, there may be an item that has an adverse effect.
[0055]
In the present embodiment, an operator who performs classification can designate one or more items that are considered necessary for the viewpoint of the classification desired by the operator. For example, when categorizing patent gazette documents, if the operator wants to focus on “the problem of the invention”, the “purpose”, “means for solving the problem”, “action”, “effect of the invention” are set. specify. Further, when it is desired to pay attention to solving means, “means for solving the problem” and “example” can be designated. When an item to be classified is designated, the document data is converted based on the designated item.
[0056]
In FIG. 5, when the operator designates the items “purpose”, “means for solving the problem”, “action”, and “effect of the invention”, only the content data of the designated item is displayed. It is an example when converted to include.
[0057]
As is apparent from the change data 502, the data corresponding to the item “purpose” is “to store screen information corresponding to the history and so on” and “to solve the problem”. The data corresponding to the item of “means of“ having display means for displaying to achieve the above-mentioned object ”and the data corresponding to the item of“ action ”“ in the above configuration, From the input procedure, it operates as if it is displayed, and the data corresponding to the item of “Effects of the invention” is “According to the Japanese invention as described above… there is an effect that can be reproduced”. Are combined to form one document.
[0058]
FIG. 6 shows the specified items when the operator designates “purpose”, “means for solving the problem”, “action”, and “effect of the invention” using the same document data. This is an example in which conversion is performed so that a separator 601 (“/”) is inserted between the data of each item.
[0059]
Next, a series of processing procedures of the document classification apparatus according to the first embodiment will be described. FIG. 7 is a flowchart showing a series of processing procedures of the document classification apparatus according to the first embodiment.
[0060]
In the flowchart of FIG. 7, first, the input unit 1 inputs document data (step S710). The designation unit 402 designates an item (step S720).
[0061]
The conversion unit 403 converts the document data input in step S710 into converted data so as to be the contents of the item specified in step S720 (step S730). Further, the separator 601 is inserted between the data corresponding to the items as necessary (step S740). The converted conversion data is stored in the conversion data storage unit 404 together with the separation symbol data (step S750).
[0062]
Based on the converted data converted in step S730 or the converted data stored in the converted data storage unit 404 in step S750, the classification unit 405 performs document classification (step S780). After the classification process is completed, the result of the classification process is stored in the classification result storage unit 406 (step S790), and all the processes are completed.
[0063]
As described above, according to the first embodiment, document data is converted into converted data by a specified item, and the document is classified based on the converted data. Therefore, the classification based on the contents of other unnecessary items is performed. The influence on the result can be suppressed. In addition, by inserting the separation symbol 601, it is possible to identify data for each combined item in the converted data, and to avoid content confusion due to data combination between items.
[0064]
(Embodiment 2)
In the first embodiment, the documents are classified using the conversion data. However, as in the second embodiment described below, a feature vector of the document is generated using the conversion data, and the feature vector is used. You may make it classify | categorize a document.
[0065]
First, a functional configuration of the document classification apparatus according to the second embodiment will be described. FIG. 8 is a block diagram functionally showing the configuration of the document classification apparatus according to the second embodiment. In FIG. 8, the same components as those in FIG. 4 of the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.
[0066]
In FIG. 8, the document classification apparatus includes a document vector generation unit 801 in addition to an input unit 401, a specification unit 402, a conversion unit 403, a conversion data storage unit 404, a classification unit 405, and a classification result storage unit 406. And a document vector storage unit 802.
[0067]
The document vector generation unit 801 and the document vector storage unit 802 are described in a program recorded in a recording medium such as the ROM 202 or 302, the RAM 203 or 303, the disk device 306, or the hard disk 316, as with other components. The function of each unit is realized by the CPU 201 or 301 executing instruction processing in accordance with the received instructions.
[0068]
(Document vector generation unit 801)
The document vector generation unit 801 generates a feature vector for each document. In order to generate a feature vector of a document, it is necessary to perform natural language analysis processing such as morphological analysis on document data. This natural language analysis processing is performed for each item of each document data by a document analysis unit (not shown). The morpheme analysis can use a conventional morpheme analysis method.
[0069]
The document vector generation unit 801 generates a document vector for each document data using the analysis result obtained by the document analysis unit. At this time, the document vector is generated based only on the analysis result relating to the item designated by the designation unit 402. For example, only the feature data obtained from the content data of the item specified by the specification unit 402 is added to each document data to generate a document vector, so that only the content data of the item specified by the specification unit 402 is reflected. Document vectors can be generated.
[0070]
(Document vector storage unit 802)
The document vector storage unit 802 is a storage unit that stores the feature vector of each document generated by the document vector generation unit 801. In the document vector storage unit 802, even if the document is the same, the feature vector of the document differs depending on the item specified by the specification unit 402. Therefore, the feature vector of the document is stored for each specification. When classifying documents by the classification unit 405, the document feature vectors stored in advance by the document vector storage unit 802 are used, so that the documents can be classified efficiently.
[0071]
As the document vector storage unit 802, for example, the disk device 206 of the server 101, the hard disk 305 on the client side, the floppy disk 307, or the like can be set according to the difference in conversion data capacity or application. .
[0072]
(Classification unit 405)
The classification unit 405 classifies documents based on the similarity between feature vectors of each document converted by the conversion unit 403. Specifically, document classification can be performed by applying classification methods such as a chi-square method, a discriminant analysis method, and a cluster analysis method to the generated classification target data. Here, any classification method can be used as long as the vector data can be applied.
[0073]
Next, a series of processing procedures of the document classification apparatus according to the second embodiment will be described. FIG. 9 is a flowchart showing a series of processing procedures of the document classification apparatus according to the second embodiment. In the flowchart of FIG. 9, steps S710 to S750 are the same as those in the flowchart of FIG. 7 of the first embodiment, and therefore the same step numbers are given and description thereof is omitted.
[0074]
Based on the converted data converted in step S730 or the converted data stored in step S750, the document vector generation unit 801 generates a feature vector of each document (step S760). The generated feature vector of each document is stored in the document vector storage unit 802 (step S770).
[0075]
Based on the conversion data converted in step S760 or the conversion data stored in step S770, the classification unit 405 performs document classification (step S780). After the classification process is completed, the result of the classification process is stored in the classification result storage unit 406 (step S790), and all the processes are completed.
[0076]
As described above, according to the second embodiment, the document data is converted into the conversion data by the designated item, and the feature vector of each document is generated based on the conversion data. Therefore, the document more reflecting the operator's intention. The feature vector can be used to classify documents, and the influence on the classification result due to the contents of other unnecessary items can be suppressed.
[0077]
【The invention's effect】
As explained above, This According to the invention, when classifying the document, only the content data of the designated item is used, so that it is possible to prevent the content of other items from affecting the classification result. Therefore, it is possible to obtain a document classification device capable of efficiently performing classification with high accuracy close to the classification desired by the operator by designating items of document data that are considered necessary for the viewpoint of the classification expected by the operator. There is an effect.
[0078]
Also, This According to the invention, when generating the feature vector of each document for classifying the document, only the content data of the designated item is used, so that the influence of the content of the other item on the classification result is prevented. Can do. Therefore, by specifying the document data items that are considered necessary for the viewpoint of the classification expected by the operator, it is possible to obtain a document classification apparatus capable of performing classification with high accuracy close to the classification desired by the operator. Play.
[0079]
Also, This According to the invention, since the delimiter symbol is inserted between the conversion data, it is possible to avoid the confusion of the data of a plurality of items as one data in the analysis process such as morphological analysis. In addition, there is an effect that a document classification device capable of instantaneously identifying the content data for each item is obtained.
[0080]
Also, This According to the invention, when classifying the document, only the content data of the designated item is used, so that it is possible to prevent the content of other items from affecting the classification result. Therefore, it is possible to obtain a document classification method capable of performing classification with high accuracy close to the classification desired by the operator by designating the items of the document data that are considered necessary for the viewpoint of the classification expected by the operator. There is an effect.
[0081]
Also, This According to the invention, when generating the feature vector of each document for classifying the document, only the content data of the designated item is used, so that the influence of the content of the other item on the classification result is prevented. Can do. Therefore, it is possible to obtain a document classification method capable of performing classification with high accuracy close to the classification desired by the operator by designating the items of the document data that are considered necessary for the viewpoint of the classification expected by the operator. There is an effect.
[0082]
Also, This According to the invention, since the delimiter symbol is inserted between the conversion data, it is possible to avoid the confusion of the data of a plurality of items as one data in the analysis process such as morphological analysis. In addition, there is an effect that it is possible to obtain a document classification method capable of instantaneously identifying content data for each item.
[0083]
Also, This According to the invention of the above By recording a program that causes a computer to execute the method described in any one of the above, the program can be machine-readable, the above There is an effect that a recording medium capable of realizing the above operation by a computer can be obtained.
[Brief description of the drawings]
FIG. 1 is an explanatory diagram showing a hardware configuration of an entire information processing system that constitutes a document classification device according to Embodiment 1 of the present invention;
FIG. 2 is an explanatory diagram showing a server in hardware in the information processing system constituting the document classification device according to the first embodiment.
FIG. 3 is an explanatory diagram showing a client in the information processing system constituting the document classification device according to the first embodiment in hardware.
FIG. 4 is a block diagram functionally showing the configuration of the document classification device according to the first embodiment.
FIG. 5 is an explanatory diagram showing an example of the contents of document data and conversion data in the document classification device according to the first embodiment.
6 is an explanatory diagram showing another example of the contents of document data and converted data in the document classification device according to Embodiment 1. FIG.
FIG. 7 is a flowchart showing a series of processing procedures of the document classification device according to the first embodiment.
FIG. 8 is a block diagram functionally showing the configuration of a document classification apparatus according to Embodiment 2 of the present invention.
FIG. 9 is a flowchart showing a sequence of processing steps of the document classification device according to the second embodiment.
[Explanation of symbols]
101 server
102 clients
103 network
201 CPU
204 I / F
206 Disk unit
301 CPU
306 hard disk
308 display
309 I / F
311 keyboard
312 mouse
313 Scanner
401 Input section
402 Designation part
403 conversion unit
404 Conversion data storage unit
405 Classification part
406 Classification result storage unit
501 Document data
502 Conversion data
601 Separation symbol
801 Document vector generation unit
802 Document vector storage unit

Claims (11)

In a document classification device that classifies documents based on document contents,
The document classification device includes :
Means for obtaining the document data,
Means for retrieving a predetermined code from the acquired document data and extracting one or more items;
Means for causing specified at least one of the extracted items,
Means for converting the document data so that only the data corresponding to the specified item is contained;
Means for classifying the document using the converted data;
Have
The means for converting is
An apparatus for classifying documents, wherein a separator is inserted between the data of the items when the data corresponding to the designated items does not form a sentence. In a document classification device that classifies documents based on document contents,
The document classification device includes :
Means for obtaining the document data,
Means for retrieving a predetermined code from the acquired document data and extracting one or more items;
Means for causing specified at least one of the extracted items,
Means for converting the document data so that only the data corresponding to the specified item is contained;
Means for generating a feature vector of each document using the converted data;
Means for classifying the document using the feature vector of each generated document;
Have
The means for converting is
An apparatus for classifying documents, wherein a separator is inserted between the data of the items when the data corresponding to the designated items does not form a sentence.   3. The document classification apparatus according to claim 1, wherein a symbol that does not exist in the conversion data is used as the separation symbol. The document classification apparatus according to claim 1, wherein the conversion unit can be set by an operator as to whether or not to insert the separation symbol. The document classification apparatus according to claim 1, wherein the document data has one or more predetermined codes. The designated means is:
  Display means for displaying items extracted by the extracting means;
  Means for designating at least one item from among the items presented on the display means;
The document classification apparatus according to claim 1, comprising: In a document classification method in a document classification device that classifies a document based on the contents of the document,
A step performed by the document classification device,
An acquisition unit included in the document classification apparatus acquires document data composed of one or more items;
An extraction unit included in the document classification device , searching the item from the acquired document data and extracting one or more items;
A designation means provided in the document classification device designates at least one or more of the extracted items;
A step of converting the document data so that the conversion means included in the document classification apparatus has only the data corresponding to the designated item;
A classifying unit provided in the document classification device classifies the document using the converted data;
Have
The converting step includes
A document classification method, wherein a separator is inserted between data of items when the data corresponding to the designated items does not form a sentence. In a document classification method in a document classification device that classifies a document based on the contents of the document,
A step performed by the document classification device,
An acquisition unit included in the document classification apparatus acquires document data composed of one or more items;
An extraction unit included in the document classification device , searching the item from the acquired document data and extracting one or more items;
A designation means provided in the document classification device designates at least one or more of the extracted items;
A step of converting the document data so that the conversion means included in the document classification apparatus has only the data corresponding to the designated item;
Generation means for the document classification apparatus comprising, generating a feature vector for each document by using the converted data,
A classifying unit provided in the document classification apparatus classifies the document using the feature vector of each generated document;
Have
The converting step includes
A document classification method, wherein a separator is inserted between data of items when the data corresponding to the designated items does not form a sentence. 9. The document classification method according to claim 7 , wherein a symbol that does not exist in the conversion data is used as the separation symbol. 9. The document classification method according to claim 7 , wherein the converting step can be set by a user as to whether to insert the separator. A computer-readable recording medium having recorded thereon a program for causing the document classification apparatus to execute each step of the method according to any one of claims 7 to 10 .

Images