JPH11250067A - Information classification method and its device, and medium for storing information classification program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the precision of keyword extraction, to facilitate information management by a service user and to efficiently collect information at a high speed. SOLUTION: In an information classification method, a general keyword is extracted from information at the time of registration or registered information, an information category is classified based on the general keyword, general classification information is stored, an organization keyword storage means accumulating the keyword peculiar to an organization is referred to, an organization keyword is extracted from information at the time of registration or registered information, an information category is classified based on the extracted organization keyword, organization classification information is stored and general classification information and organization classification information, which are stored, are arranged on a two dimensional plane.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information classification method and apparatus, and a storage medium storing an information classification program. You can register arbitrarily,
When searching for necessary information, the registered information is classified according to the purpose and means, and the classification status is visualized and presented to the user, so that the user can manage and collect the information. The present invention relates to an information classification method and apparatus for facilitating information storage and a storage medium storing an information classification program.

[0002]

2. Description of the Related Art As information management in a conventional information management service, all registered information is handled in a unified manner, and classification is not performed. When a registrant or an information management service administrator registers information, Classify into appropriate category: Automatically classify information by extracting keywords included in the information from registered information, etc .:

[0003] The information classification method described above will be described below. FIG. 7 shows a configuration of a conventional information classification device. The information classification device shown in FIG. 1 includes a control unit 100, a network interface unit 130, a reception unit 140, a transmission unit 150, a service processing unit 200, an information storage unit 210,
Keyword extraction unit 310, general category classification unit 320,
The network interface 130 is connected to a network 120, and the network 120 includes a plurality of user terminals 11
0 is connected.

In order to activate the information classification apparatus shown in FIG. 1 and operate the present service, first, a service user who refers to information from the user terminal 110, issues a command for posting, and a text of the information to be posted. The input data is transmitted via the network 120 according to an instruction from the control unit 100.
The data is read into the network interface unit 130. After the input data is input, the network interface unit 130 transfers the input data to the receiving unit 140 according to an instruction of the control unit 100.

[0005] After receiving the input data, the receiving section 140 transfers the input data to the information storage section 210 according to an instruction from the control section 100. After acquiring the input data, the information storage unit 210 stores the input data in the control unit 100.
The command included in the stored input data is stored according to the instruction of the
Transfer to 0.

[0006] After acquiring the command, the service processing unit 200 analyzes the content of the command in accordance with an instruction from the control unit 100 and instructs the information storage unit 210 to process data for executing the process. After acquiring the instruction data from the service processing unit 200, the information storage unit 210 transfers the processing data indicated by the instruction data to the service processing unit 200 according to an instruction from the control unit 100.

[0007] After acquiring the processing data, the service processing section 200 performs the processing in accordance with an instruction from the control section 100.
When the processing content in the service processing unit 200 is “registration” of information, the information to be registered in the information storage unit 210 is transferred and the information is transferred to the keyword extraction unit 310 according to an instruction from the control unit 100. After acquiring the information to be registered, the keyword extracting unit 310 extracts a keyword necessary for the category classification from the information acquired according to an instruction of the control unit 100, and transfers the extracted keyword to the category classification unit 320.

After the category classification unit 320 obtains, the control unit 1
The information is classified according to the instruction of 00, and the classified result is transferred to the classification information storage unit 330. Classification information storage unit 330
After the acquisition, the classification result is stored in accordance with an instruction from the control unit 100. If the processing content in the service processing unit 200 is information retrieval, the control unit 100 inquires of the classification information storage unit 330 about the category to be retrieved.

After the acquisition, the classification information storage unit 330 transfers a list of information belonging to the search target category to the service processing unit 200 according to an instruction from the control unit 100. After the processing is completed, the service processing unit 200 transfers the processing result data to the transmission unit 150 according to an instruction of the control unit 100.
After acquiring the processing result data, the transmitting unit 150 outputs the processing result data to the network interface unit 130 according to an instruction of the control unit 100.

[0010] The network interface unit 130
After acquiring the processing result data, the processing result is output to the user terminal 110 via the network 120 according to the instruction of the control unit 100. From the above, the service user,
When posting information to the information management service, information is automatically classified in the service, so that information management and information retrieval at the time of information collection are facilitated.

Next, a conventional keyword extracting method in the keyword extracting section 310 will be described. As a conventional keyword extraction method, there is the following method. At the time of information registration, the information registrant assigns a keyword to the information by himself / herself: At the time of information registration, a list of keywords prepared in advance in the present service is presented to the information registrant and made to select: Analyze the text, and if a keyword that matches the keyword list prepared in advance in this service is included, use that keyword: When registering information, analyze the information text, extract words, and The evaluation points are given using the appearance frequency and the appearance order of, and the evaluation points exceed a certain threshold, or
A keyword that satisfies a certain criterion, such as a certain number of words having a high evaluation score, is adopted as a keyword. Further, a conventional category classification method in the general category classification unit 320 will be described. As a conventional category classification method, there is the following method. At the time of information registration, the information registrant assigns a category to information by himself / herself: At the time of information registration, a list of categories prepared in advance in the information processing service is presented to the information registrant for selection: The registered keyword is compared with a keyword category correspondence table prepared in advance, and a matching category is adopted. At the time of information registration, the registered keyword is checked against an automatically generated keyword-category correspondence table, and a matching category is adopted.

As an example of the automatic generation method, a keyword group having a high degree of relevance among all keywords registered in the information processing service is determined based on the frequency of occurrence of a plurality of keywords assigned to information. Can be considered as one category, a category name is given to each of the generated categories, and a keyword-category correspondence table is created.

[0013]

However, in the above-described conventional information management method, it is difficult for an information management service user to collect necessary information because the category to which the information belongs is unknown. Further, it is also difficult to grasp the tendency of the field of information registered in the service.

Further, in the above-mentioned conventional information management method, there is a problem that the cost of information management becomes extremely large because information classification is performed manually. In addition, in the method shown in FIG. 7, since the keyword is unifiedly extracted regardless of the use of the keyword extracted in the keyword extracting unit 310, a keyword which is specially used in the organization using the service is used. May not be extracted properly. Furthermore, the case where the extracted keyword indicates the field and purpose of the information and the case where the extracted keyword indicates the means or action that acquired the information are treated equally, so that it is not easy to narrow down the information from both the purpose and the means. And the like. It can be easily inferred that this problem occurs regardless of the algorithm or accuracy of keyword extraction.

As described above, when a keyword is extracted from information for the purpose of category classification, the number of appearances and the appearance position of the keyword included in the information are important factors.
Used in many keyword extraction methods. On the other hand,
The information handled in an organization includes keywords that are unique to the organization, such as specific proper nouns and abbreviations.

On the other hand, in an information management service used by an organization, a lot of information closely related to an organization-specific event or event is registered and referred to. Therefore, the keyword indicating the means by which information was obtained or the event for which information acquisition was permitted,
It is important regardless of the number of appearances and the appearance position in the information body. In the category classification method in the conventional information processing service management device, since keywords are uniformly extracted from each information to classify the information, an important specific keyword may not be extracted in the organization. However, there is a drawback in that information cannot be classified from a viewpoint according to the event.

SUMMARY OF THE INVENTION The present invention has been made in view of the above points, and improves the accuracy of keyword extraction, facilitates management of information by a service user, and performs high-speed and efficient information collection. It is an object of the present invention to provide a method and an apparatus for classifying information and a storage medium storing an information classification program.

[0018]

FIG. 1 is a diagram for explaining the principle of the present invention. The present invention (claim 1) provides an information management service for managing a group of collected information, an information classification method for classifying information according to purpose or means, wherein information at the time of registration or registered information is used. A general keyword is extracted (step 1), an information category is classified based on the extracted general keyword, the classification information is stored (step 2), and an organization that stores an organization-specific keyword is stored. Referring to the keyword storage means, an organization keyword is extracted from the information at the time of registration or from the registered information (step 3), the information category is classified based on the extracted organization keyword, and the classification information is stored. In advance (step 4), the stored general classification information and organization classification information are arranged two-dimensionally (step 5).

FIG. 2 is a diagram showing the principle of the present invention. The present invention (claim 2) provides an information classification service for classifying information according to purpose and means in an information management service for managing a collected information group, wherein an organization keyword storage for accumulating keywords specific to an organization. Means 10, general keyword extracting means 20 for extracting general keywords from information at the time of registration or registered information, and information categories are classified based on general keywords extracted by general keyword extracting means 20 The information at the time of registration or the registered information is referred to by referring to the general category classification means 30, the general classification information storage means 40 for storing the information classified by the general category classification means 30, and the organization keyword storage means 10. Keyword extracting means 50 for extracting a keyword from an organization, and keyword extracting means 50
An organization category classification unit 60 that classifies information categories based on the organization keywords extracted by the above, an organization classification information storage unit 70 that stores information classified by the organization category classification unit 60, a general classification information storage unit 40,
There is provided a visualization arrangement unit 80 for arranging the general classification information and the organization classification information stored in the organization classification information storage unit 70 in a two-dimensional manner.

According to the present invention (claim 3), an organization-specific keyword is stored in the organization keyword storage means 10 as an organization keyword based on the results of a questionnaire in a certain organization. According to the present invention (claim 4), in the organization keyword storage means 10, keywords extracted based on proper nouns or information acquisition means / events among keywords extracted as general keywords are stored as organization keywords.

According to the present invention (claim 5), the organization keyword storage means 10 analyzes the use history of other services used in a certain organization, and uses keywords frequently used for information retrieval as organization keywords. Store. According to the present invention (claim 6), the organization category classification means 60 includes a means for referring to an organization keyword-category correspondence table prepared in advance.

According to a seventh aspect of the present invention, in an information management service for managing a group of collected information, a storage medium storing an information classification program for classifying information according to a purpose or a process is provided. Or a general keyword extraction process of extracting general keywords from registered information, and an information category is classified based on the general keywords extracted by the general keyword extraction process. A general category classification process for storing information, an organization keyword extraction process for extracting organization keywords from information at the time of registration or registered information with reference to an organization keyword storage means for storing organization-specific keywords, Information categories based on organization keywords extracted by the keyword extraction process Classification and has a tissue category classification process for storing the classified tissue classification information, and a visualization arrangement process to arrange the general classification information and tissue classification information stored on a two-dimensional.

As described above, according to the present invention, in order to efficiently extract keywords from information and to perform information classification in a form suitable for the use form of the organization, two keywords are extracted. Divide into processing. On the one hand, keyword extraction and category classification are performed based on the conventional method, and on the other hand, only the keywords unique to the organization and the means for acquiring information and the keywords based on the events are extracted and classified.

Further, by arranging the information on the two-dimensional screen using the information categories classified by both methods, the distribution of the information registered in the information management service is visualized and provided to the service user. . This allows
By dividing the keyword extraction process for category classification, keywords unique to the organization can be efficiently extracted, and by arranging information two-dimensionally together with the category classification result by the conventional method, The information management by the service user becomes easy, and the information can be collected quickly and efficiently.

[0025]

FIG. 3 shows the configuration of an information classification device according to the present invention. The information classification device shown in the figure includes a network interface unit 130, a receiving unit 140, a transmitting unit 15
0, control section 600, service processing section 700, information storage section 710, general keyword extraction section 810, general category classification section 820, general classification information storage section 830, organization keyword storage section 900, organization keyword extraction section 910, organization category classification A network interface unit 130 is connected to the network 120 to which the plurality of user terminals 110 are connected. The unit 920 includes an organization classification information storage unit 930 and a category visualization unit 1000.

A plurality of user terminals 110 use the service via the network 120. In the following description, an example in which the user terminal 110 accesses the apparatus will be described. The same procedure is used when using from another user terminal, and the present invention can be similarly applied when the number of user terminals is three or more.

The network interface unit 130
The communication with the user terminal 110 is performed via the network 120. The receiving unit 140 acquires information from the network interface unit 130. The transmission unit 150 transfers the information acquired from the service processing unit 700 and the information storage unit 710 via the network interface unit 130.

The control section 600 controls all the components of the information classification device. The service processing unit 700 analyzes an instruction from the control unit 600 and transfers processing data for executing the process to a necessary processing unit. Information storage unit 71
0 reads out the information instructed from the service processing unit 700 and passes it to the service processing unit 700.

The general keyword extracting section 810 extracts general keywords necessary for category classification based on information obtained from the service processing section 700. The general category classification unit 820 classifies information categories based on the keywords extracted by the general keyword extraction unit 810. The general classification information storage unit 830 stores the classification status of the general category classified by the general category classification unit 820.

The organization keyword extraction section 910 extracts a keyword unique to the organization from each information. In order to extract the keyword unique to the organization, the following three methods can be considered. Take a questionnaire within the organization that uses the service,
Select keywords specific to your organization.

Of the keywords extracted as general keywords, only proper nouns, information acquisition means, and keywords based on events are extracted and registered as organization keywords. The use history of other services used by the organization using this service is analyzed, keywords frequently used for information retrieval are extracted, and keywords based on proper nouns, information acquisition means, and events are registered as organization keywords.

The organization category classification section 920 classifies information categories based on the organization keywords extracted by the organization keyword extraction section 910. Organization classification information storage unit 93
0 stores the classification status of the organization category separated by the organization category classification unit 920. Category visualization unit 1000
The general classification information storage unit 830 and the organization classification information storage 93
Based on the classification information of the category stored in 0, they are arranged two-dimensionally.

In the above configuration, the general category classification unit 8
20 and the category classification process performed by the organization category classification unit 920 uses the method described in and in the above-described conventional technique. However, the keyword-
The category correspondence table targets only those registered as organization keywords.

[0034]

Embodiments of the present invention will be described below with reference to the drawings. First, the operation of the information classification device having the above configuration will be described below. FIG. 4 is a flowchart showing the operation of the information classification device according to one embodiment of the present invention.

Step 101) In order to operate the service, first, a command for referring to or posting information from the user terminal 110 and data input by the service user such as a text of the information to be posted are transmitted from the control unit 600 to the control unit 600. It is read into the network interface unit 130 via the network 120 according to the instruction. After the input data is input, the network interface unit 130
The input data is received by the receiving unit 14 according to an instruction from the control unit 600.
Transfer to 0.

After receiving the input data, receiving section 140 transfers the input data to information storage section 710 according to an instruction from control section 600. Step 102) After acquiring the input data, the information storage unit 710 stores the input data according to an instruction from the control unit 600. The information storage unit 710 transfers the command included in the input data to the service processing unit 700 according to an instruction from the control unit 600.

After acquiring the command, the service processing unit 700 analyzes the content of the command according to an instruction of the control unit 600, and instructs the information storage unit 710 to execute processing data for executing the process. After acquiring the instruction data from the service processing unit 700, the information storage unit 710 transfers the processing data indicated by the instruction data to the service processing unit 700 according to an instruction from the control unit 600.

Step 103) Service processing section 700
Performs the process according to the instruction of the control unit 600 after acquiring the process data. It is determined whether the processing content in the service processing unit 700 is “registration” of the information. If the content is “registration”, the process proceeds to step 104; otherwise, the process proceeds to step 109.

Step 104) In accordance with an instruction from the control section 600, information to be registered in the information storage section 710 is transferred, and the information is transferred to the general keyword extraction section 81 and the organization keyword extraction section 910. Step 105) After acquiring the information, the general keyword extracting unit 810 extracts keywords necessary for category classification from the acquired information according to an instruction of the control unit 600, and extracts the extracted keywords from the general category classifying unit 820.
Transfer to

Step 106) General category classification unit 8
20 obtains the keyword from the general keyword extraction unit 810, classifies the general category according to the instruction of the control unit 600, transfers the classified result to the general classification information storage unit 830, and performs classification in the general classification information storage unit 830. The result is stored. Step 107) After acquiring the information, the organization keyword extraction unit 910 transfers a command requesting a list of organization keywords from the organization keyword storage unit 900 according to the instruction of the control unit 600.

After acquiring the list of organization keywords, the organization keyword storage unit 900 transfers the list of organization keywords registered in advance to the organization keyword extraction unit 910 according to an instruction from the control unit 600. Organization keyword extraction unit 91
0 is obtained from the list of the organization keywords, and the control unit 600
, The information is matched with the organization keyword,
The matching keyword is transferred to the organization classification category classification unit 920.

Step 108) Organization category classification section 9
After obtaining the matching keyword, the control unit 600 classifies the organization category using the keyword, and transfers the classification result to the organization classification information storage unit 930. Step 109) In the above step 103,
If it is determined that the content is not “register”, it is determined whether the content of the process is “search”. If so, the process proceeds to step 110; otherwise, the process ends.

Step 110) Service processing section 700
Is a category visualization unit 100 according to an instruction from the control unit 600.
0 issues a command for acquiring a category and a list of information belonging to each category. As a result, the category visualization unit 1000 issues a command for acquiring an information list to the general classification information storage unit 830 and the organization classification information storage unit 930.

Step 111) General classification information storage section 8
30 and the organization classification information storage unit 930 transfer a category and a list of information belonging to each category to the category visualization unit 1000 according to an instruction of the control unit 600. After acquiring the list, the category visualization unit 1000 arranges the items on the two-dimensional screen using the items indicating the general category and the items indicating the organization category according to the instruction of the control unit 600, and transfers the result to the service processing unit 700. . Thus, the service processing unit 700 transfers the processing result data to the transmission unit 150 according to the instruction of the control unit 600 after the processing is completed. After acquiring the processing result data, the transmitting unit 150 outputs the processing result data to the network interface unit 130 according to an instruction of the control unit 600. After acquiring the processing result data, the network interface unit 130 outputs the processing result to the user terminal 110 via the network 120 according to the instruction of the control unit 600.

FIG. 5 shows an example of the classification result of one embodiment of the present invention. The example shown in the figure is an example of general keywords, general categories, organization keywords and organization categories extracted and classified by the above processing. The general keywords and general categories are stored in the general classification information storage unit 830, The keyword and the organization category are stored in the organization classification information storage unit 8
30.

In the example shown in FIG.
At 0, "BASIC", "FORTRAN",
It can be seen that “Perl” is extracted, and “program language” is extracted as a category corresponding to the keyword in the general category classification unit 820. In addition, since “organizational keyword”, “meeting”, and “tour” are present in the organization keyword storage unit 900 in the organization keyword extraction unit 910, “business trip report” is extracted as a category by the keyword in the organization category classification unit 920. Things.

It can be easily analogized that the contents of the organization keyword and the organization category differ depending on the organization to which the present apparatus is applied. These keywords and categories are
These are stored in the general classification information storage unit 830 and the organization classification information storage unit 930, respectively, and are processed as “search”
Is transferred, the category visualization unit 1000 reads the information from the storage units 830 and 930, converts the information into a displayable form as shown in FIG. 6, and transfers the information to the user terminal 110 via the network 120. I do. Thereby, the user terminal 110 can display the classification result shown in FIG. Note that the category visualization unit 1000 converts the classification information, which is text information, into a displayable form by an existing method so that the general classification information and the organization classification information can be visualized. In the example shown in FIG.
The relationship between the two categories is expressed by changing the size of the circle in the matrix according to the similarity between the keywords included in the general classification information and the category in the organization classification information.

It should be noted that the information set in FIG. 6 is an example, and is not particularly limited to the visualization example.
Numerical values and various display and editing methods are possible. In the above embodiment, the description has been given based on the components shown in FIG. 3. However, the present invention is not limited to this example. Each component is constructed as a program, and a disk connected to a computer functioning as an information classification device. The present invention can be easily realized by storing it in a device, a portable storage medium such as a floppy disk or a CD-ROM, and installing it when implementing the present invention.

The present invention is not limited to the above embodiment, but can be variously modified and applied within the scope of the claims.

[0050]

As described above, according to the present invention, when a user of an information management service uses an information management service to manage and collect information in an organization, a keyword unique to the organization, an event of the organization, and the like. By obtaining the information acquisition means and the keywords related to the information acquisition permission separately from general keywords, the accuracy of keyword extraction is improved, and the category classification results are visualized and displayed on the secondary screen. In addition, the management of information by the service user becomes easy, and the information can be collected at high speed and efficiently.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining the principle of the present invention.

FIG. 2 is a principle configuration diagram of the present invention.

FIG. 3 is a configuration diagram of an information classification device of the present invention.

FIG. 4 is a flowchart showing the operation of the information classification device according to one embodiment of the present invention.

FIG. 5 is an example of a classification result according to an embodiment of the present invention.

FIG. 6 is an example of information set in a category visualization unit according to an embodiment of the present invention.

FIG. 7 is a configuration diagram of a conventional information classification device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Organization keyword storage means 20 General keyword extraction means 30 General category classification means 40 General classification information storage means 50 Organization keyword extraction means 60 Organization category classification means 70 Organization classification information storage means 80 Visualization arrangement means 110 User terminal 120 Network 130 Network interface unit 140 Reception unit 150 Transmission unit 600 Control unit 700 Service processing unit 710 Information storage unit 810 General keyword extraction unit 820 General category classification unit 830 General classification information storage unit 910 Organization keyword extraction unit 920 Organization category classification unit 930 Organization classification information storage unit 1000 Category visualization unit

Claims (7)

[Claims] In an information management service for managing a group of collected information, in an information classification method for classifying information according to purpose or means, a general keyword based on information at the time of registration or registered information. Is extracted, the information category is classified based on the extracted general keywords, general classification information is stored, and an organization keyword storage unit that stores organization-specific keywords is referred to. Information or an organization keyword is extracted from the registered information, an information category is classified based on the extracted organization keyword, organization classification information is stored, and the stored general classification information and the An information classification method characterized by arranging organization classification information two-dimensionally. 2. An information classification service for classifying information according to purpose and means in an information management service for managing a group of collected information, comprising: an organization keyword storage means for storing an organization-specific keyword; General keyword extracting means for extracting general keywords from information at the time or registered information; general category classifying means for classifying information categories based on the general keywords extracted by the general keyword extracting means A general classification information storage unit that stores general classification information classified by the general category classification unit; and an organization keyword is extracted from information at the time of registration or registered information with reference to the organization keyword storage unit. Means for extracting organization keywords, and the organization key extracted by the organization keyword extraction means. Organization category classification means for classifying information categories based on keywords, organization classification information storage means for storing organization classification information classified by the organization category classification means, general classification information storage means, and the organization classification information storage An information classification device, comprising: a visualization arrangement unit for arranging the general classification information and the organization classification information stored in a unit in a two-dimensional manner. 3. The organization keyword storage means stores an organization-specific keyword as an organization keyword based on a result of a questionnaire in a certain organization.
Described information classification device. 4. The organization keyword storage unit stores, as an organization keyword, a keyword extracted based on a proper noun or an information acquisition unit and an event among keywords extracted as general keywords.
Described information classification device. 5. The organization keyword storage unit according to claim 2, wherein the organization keyword storage unit analyzes a use history of another service used in a certain organization, and stores a keyword frequently used for information search as an organization keyword. Information classifier. 6. The information classification apparatus according to claim 2, wherein the organization category classification unit includes a unit that refers to an organization keyword-category correspondence table prepared in advance. 7. An information management service for managing a group of collected information, which is a storage medium storing an information classification program for classifying information according to a purpose or a process, wherein the information at the time of registration or the registered information is stored. A general keyword extraction process of extracting a general keyword from information that is present, classifying an information category based on the general keyword extracted by the general keyword extraction process,
An organization that extracts the organization keyword from the information at the time of registration or the registered information by referring to the general category classification process for storing the classified general classification information and the organization keyword storage unit that stores the organization-specific keywords. A keyword extraction process; an organization category classification process for classifying information categories based on the organization keywords extracted by the organization keyword extraction process; and storing the classified organization classification information; and the stored general classification information. And a visualization arrangement process for arranging and arranging the organization classification information two-dimensionally. A storage medium storing an information classification program.