JP4807364B2 - Information management device - Google Patents

Description

  The present invention relates to an information management apparatus, and more particularly to an information management apparatus that stores and manages input data.

  In recent years, with the increase of users of the Internet, leakage of personal information due to bulletin boards, blogs, and file sharing has become a problem. Normally, when personal information is written on bulletin boards or blogs, personal information that is inappropriate for the administrator / user who manages such bulletin boards or blogs to monitor and publish data It was necessary to delete it from the server. However, because a large number of unspecified people write information on bulletin boards and blogs, administrators and users cannot fully monitor the site, and it is also difficult to monitor constantly. It is. For this reason, there was a problem that leakage of personal information occurred.

  On the other hand, Patent Document 1 discloses a technique for preventing the leakage of personal information described above. The technique disclosed in this document employs a configuration in which information is registered in an information management apparatus as it is, and personal information is masked at the time of output.

JP 2006-190041 A

  However, as described above, if information including personal information is held in the information management apparatus as it is, the information itself including personal information may unexpectedly flow out. For example, in recent years, file exchange software for exchanging files between users on a network has been used by many users, and the personal information stored in the information management apparatus is included through such software. Data may be leaked as it is. Therefore, a technique for suppressing further information leakage is required.

  For this reason, an object of the present invention is to prevent leakage of personal information and improve security, which are the problems described above.

  Therefore, one aspect of the present invention is an information management device that receives input of predetermined data, stores the input data in a storage device, and manages the input data based on preset personal information determination criteria. Personal information detecting means for detecting a personal information part in the data, and data processing for storing the personal information part detected by the personal information detecting means so that the personal information part cannot be recognized and stored in a storage device And means.

  A program according to another aspect of the present invention receives an input of predetermined data, and stores the input data in a storage device and manages the input data based on a personal information determination criterion set in advance. , Personal information detecting means for detecting the personal information part in the input data, and data in which the personal information part is detected by the personal information detecting means are processed so that the personal information part cannot be recognized and stored in the storage device. It is characterized by realizing data processing means.

  Furthermore, another aspect of the present invention is an information management method for receiving input of predetermined data, storing the input data in a storage device and managing the input data, based on a preset personal information determination criterion, Personal information detection process for detecting a personal information part in input data, and data for which the personal information part is detected in the personal information detection process and processed so that the personal information part cannot be recognized and stored in a storage device And a processing step.

  Since the present invention is configured as described above, the personal information portion of the input data is processed so that it cannot be recognized, and then the input data is stored and managed in the storage device, thereby preventing leakage of personal information. And has an excellent effect that security can be improved.

  The present invention is characterized in that an information management device that stores and manages data detects a personal information part at the time of data registration, processes the data so that it cannot be recognized, and then registers and manages the data. . In other words, the personal information portion is not processed at the time of output with respect to the managed data, but the personal information is processed and registered so that it cannot be recognized at the time of registration.

  An information management apparatus according to an aspect of the present invention is an information management apparatus that receives input of predetermined data, stores the input data in a storage device, and manages the input data. A personal information detecting means for detecting a personal information portion in input data based on the data, and processing the data in which the personal information portion is detected by the personal information detecting means so that the personal information portion cannot be recognized. And a data processing means for storing the data.

  At this time, a personal information determination dictionary for storing a preset keyword or character string pattern which is the personal information determination criterion is provided, and the personal information detection means includes a keyword or a keyword stored in the personal information determination dictionary in the input data. A part corresponding to the character string pattern is detected as a personal information part. Further, the personal information detecting means extracts a noun part in the input data based on a preset noun extraction criterion, and requests a search from the search server accessible via the network for the noun part. Based on the search result from the search server, the noun part is detected as the personal information part. For example, the personal information detection means detects the noun part as the personal information part when the number of search hits based on the search result from the search server is equal to or less than a preset threshold value. The personal information detecting means registers the personal information portion detected based on the search result from the search server in the personal information determination dictionary.

  According to the above invention, first, when predetermined data is input to the information management device, the personal information portion in the input data is detected. For example, the personal information portion in the input data is detected based on a keyword that is personal information itself prepared in advance or a character string pattern that predicts the appearance of personal information. Alternatively, the noun part in the input data is extracted, the search result of the noun part is obtained by a search engine, and the noun part is detected as a personal information part when the number of hits is equal to or less than a predetermined threshold. . When it is detected that the personal information is included in the input data, the data is processed so that the personal information portion cannot be recognized, and the input data is stored in the storage device. Therefore, if the input data is intentionally output from the information management device in response to an access request for the input data, or if the input data is unexpectedly leaked, Since the personal information portion has already been processed so that it cannot be recognized when registering in the storage device, leakage of personal information can be prevented. Therefore, the security of the information management apparatus can be improved.

  In addition, the information management device includes a data write stop unit that stops writing the input data into the storage device until the processing by the personal information detection unit and the data processing unit with respect to the input data is completed. . The data writing stop means generates identification data having the same identification information as the identification information for specifying the input data and having a data content different from that of the input data. A configuration is adopted in which dummy data is deleted after completion of processing by the data processing means. Further, the data writing stop means adopts a configuration in which the identification information specifying the input data is changed to other identification information until the processing by the personal information detection means and the data processing means is completed.

  As a result, the input data is not written to the storage device until the processing for making the personal information portion unrecognizable as described above is completed, so that leakage of the personal information portion can be more reliably suppressed. . In particular, by generating dummy data having the same identification information as the input data or changing the identification information of the input data, it is possible to more reliably access the input data itself that has not undergone data processing on the personal information portion. Therefore, it is possible to further improve security.

  An information management method for receiving input of predetermined data and storing and managing the input data in a storage device, which is another embodiment of the present invention, is based on a preset personal information determination criterion. A personal information detection step for detecting a personal information portion in the data, and data processing for processing the data in which the personal information portion is detected in the personal information detection step so that the personal information portion cannot be recognized and stored in a storage device And a process.

  And the said personal information detection process employ | adopts the structure of detecting the part corresponding to the keyword or character string pattern which is a preset personal information criteria in input data as a personal information part. In addition, the personal information detecting step extracts a noun part in the input data based on a preset noun extraction criterion, and requests a search from the search server accessible via the network for the noun part. The noun part is detected as the personal information part based on the search result from the search server. Furthermore, the information management method employs a configuration in which a data writing stop process is provided in which writing of the input data to the storage device is stopped until the processing by the personal information detection process and the data processing process for the input data is completed.

  Hereinafter, specific configurations and operations of the present invention will be described in the embodiments. In the embodiment, the case where data is registered in a file format from a client terminal is illustrated, but the present invention is not limited to being applied to data input in a file format. For example, the present invention can also be applied to data that is not in a file format when inputting data written to a bulletin board or blog. In such a case, the written input data itself or registration data based on the input data may be handled as one file or record on the information management apparatus side.

<Embodiment 1>
A first embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing the configuration of the entire network to which file servers are connected. FIG. 2 is a functional block diagram showing the configuration of the file server. FIG. 3 is a diagram illustrating a change in input data processed by the file server. 4 to 6 are flowcharts showing the operation of the file server.

[Constitution]
As shown in FIG. 1, a file server 1 that is an information management apparatus in this embodiment includes client terminals 21 and 22 that register and use files via a network N, and a search server 3 that establishes a search site. And connected to.

  FIG. 2 shows the configuration of the file server 1. The file server 1 in the present embodiment is a general server computer including a CPU that is an arithmetic device and a flash memory and a hard disk device that are storage devices. And the file system 10 provided with the data input reception part 11, the personal information masking module 12, and the data output part 16 constructed | assembled by incorporating a predetermined program in CPU is provided. In addition, a file database 18 and a personal information masking dictionary 17 are formed in the storage device. Then, the file system 10 and the file database 18 store and manage the data input from the client terminal 21 in the file database 18, and the information in the file database 18 is changed according to the request from the client terminal 22. providing.

  In particular, the file server 1 in the present embodiment has the personal information masking module 12 described above, and when personal information exists in the input data to be registered at the time of data registration in the file database 18, It has a function of processing and registering personal information so that it cannot be recognized. Specifically, the personal information masking module 12 includes a file write lock unit 13, a personal information detection unit 14, and a personal information masking unit 15, as shown in FIG.

  The file write lock unit 13 (data write stop unit) first locks to stop writing to the file database 18 of the input file, which is input data received by the data input receiving unit 11 from the client terminal 21. To do. At this time, the file write lock unit 13 adds a unique extension “.check” to the end of the file name (identification information) of the input data indicated by reference numeral D1 in FIG. 3 as indicated by reference numeral D2 in FIG. To convert the file name. Further, the file write lock unit 13 generates a dummy file having the same file name (identification information) as the input data and different data contents, as indicated by reference numeral D3 in FIG. At this time, the content of the dummy file is empty, for example.

  Then, as shown in FIG. 3, the dummy file is deleted by the file write lock unit 13 when processing by the personal information detection unit 14 and personal information masking unit 15 described later is completed. Similarly to the above, the file name of the input data is as shown by reference numeral D4 in FIG. 3 when processing by the personal information detection unit 14 and personal information masking unit 15 described later is completed by the file write lock unit 13. The file name is changed back to the original name. At this time, if there is a personal information portion in the input data D4, as will be described later, the personal information portion (the symbol d in the symbol D4 in FIG. 3) is deleted or converted into other information, and recognition is not possible. It will be masked as possible.

  Then, when the processing by the personal information detection unit 14 and the personal information masking unit 15 to be described later ends, the file write lock unit 13 cancels the stopped state of writing the input data to the file database 18. That is, as will be described later, after the data processing is performed on the personal information portion of the input data, registration into the file database 18 becomes possible.

  At this time, even if the input data itself input from the client terminal 21 is not in the file format, the same processing as described above is executed for the identification information for identifying the input data. For example, if the input data is an article for a bulletin board or a blog, the public address for accessing the public data when the received input data is converted into public data is used as the input data. The identification information to be identified is handled in the same manner as the file name described above. Then, the file server 1 that has received an article such as a bulletin board or a blog changes the information for accessing the article by adding the above-described extension to the end of the public address given to the article. . Also, a dummy article (dummy data) having the same address as the public address for the article is generated.

  Next, the personal information search unit 14 will be described. The personal information search unit 14 (personal information detection means) searches whether or not personal information is included in the input data described above, and if it is included, specifies the personal information part. Specifically, the personal information part is searched by the following two methods. First, referring to a personal information masking dictionary 17 (personal information determination dictionary) in which a keyword for specifying personal information set in advance and character string pattern information are stored, the corresponding keyword is stored. In addition, a portion that matches the character string pattern information is specified as personal information. Here, the personal information masking dictionary 17 is, for example, “10-digit number”, “numeric string including parentheses and hyphens” as character string pattern information specifying a telephone number, and “˜prefecture” as a character string pattern specifying an address. ~ City "is stored. In addition, keywords such as personal names, company names, and employee information are stored. The personal information masking dictionary 17 can add character string pattern information and keywords by an operator or by the personal information detection unit 14 as described later.

  In the method of specifying the second personal information by the personal information detection unit 14, first, a noun part existing in the input data is extracted. In extracting the noun part, a Japanese dictionary, a grammar rule that defines a grammar, and the like are stored in the file server 1 in advance, and a proper noun is extracted with reference to such information. Then, a search request is made to the search server 3 on the network N for the extracted noun part, and the number of hits as the search result is received. When the number of hits is equal to or less than a preset threshold value, the noun part is determined as personal information. For example, by setting the threshold value to be compared with the number of hits to a low value such as “10”, personal information such as the name or company name of an individual who is not a celebrity can be detected.

  The personal information detection unit 14 registers the personal information detected using the search result by the search server 3 as a keyword in the personal information masking dictionary 17 as described above. Thereby, the personal information detected as described above can be detected by matching with the personal information masking dictionary 17 thereafter.

  As described above, the personal information detection unit 14 is set in advance, such as matching with a preset personal information masking dictionary 17 or detecting the extracted noun part based on the result detected by the search server 3. Operates to detect personal information from input data based on personal information criteria.

  Next, the personal information masking unit 15 will be described. The personal information masking unit 15 (data processing means) processes the personal information part in the input data detected as described above so that it cannot be recognized. For example, the personal information part is deleted from the input data, or the personal information part is converted into a symbol such as “×”. Then, the personal information masking unit 15 stores in the file server 1 the input data whose personal information part has become unrecognizable.

  When the data output unit 16 specifies a file name or the like from the client terminal 22 and receives a data request, the data output unit 16 detects the data of the file name from the file database 18 and transmits the corresponding data to the client terminal 22. . At this time, even if the requested data contains personal information, the personal information portion has already been deleted or changed so that it cannot be recognized as described above, and thus leakage of personal information is prevented. be able to. Even if there is a request for access to input data during the processing of personal information detection or data processing in the personal information detection unit 14 and personal information masking unit 15 described above, Since the file name of the input data is converted by the file write lock unit 13, the data cannot be searched. In this case, since dummy data having the same file name as the input data is searched, in a situation where the personal information may not be masked, access to the input data itself may be prevented. it can.

[Operation]
Next, the operation of the file server 1 configured as described above will be described with reference to the flowcharts of FIGS. First, the user A who operates the client terminal 21 writes input data including personal information (see reference numeral D1 in FIG. 3) to the file server 1 (step S1). When the file server 1 tries to write the received input data to the file database 18, writing to the file database 18 is locked (step S2, data writing stop process). That is, writing of input data that has been input is stopped. At this time, the extension of “.check” is added to the file name of the input data itself to change the file name (see step S3, symbol D2 in FIG. 3), and the same file name and the data content are empty. (See step S4, reference D3 in FIG. 3). Thereafter, a process for detecting personal information in the input data whose file name has been changed is performed (step S5).

  Specifically, the personal information detection process is first performed by matching characters in the input data with keywords and character string patterns in the personal information masking dictionary 17 (step S6). Then, the matched character portion is specified as the personal information portion (Yes in step S7, personal information detection step). After that, as indicated by reference numeral D4 in FIG. 3, the personal information part d in the input data is deleted or converted into another symbol or the like, and the personal information part is masked so that it cannot be recognized (step S8). Data processing process).

  The personal information detection process is also executed by the following method. First, a noun part that does not match the personal information matching dictionary 17 is extracted from the input data (No in step S7, yes in step S13, yes in step S14), and such noun part is searched by the search server 3 on the network N. A request is made (step S15). Then, the hit number information as the search result is acquired, and the hit number is compared with a preset threshold value (step S16). At this time, if the number of hits is equal to or less than the threshold value (Yes in step S16), the searched noun part is determined as personal information (personal information detection step), and the personal information part is masked so that it cannot be recognized as described above. (Step S17, data processing step). Furthermore, as described above, the personal information portion specified based on the search result is additionally registered in the personal information matching dictionary 17 as a keyword to be handled as personal information (step S18).

  Then, the above-described detection of the personal information is performed on the entire text in the input data, and the personal information part is masked (step S9). During the above process, when an access request is made by designating the file name of the input data from another client terminal 22, a dummy file having a size of “0” is accessed. Accordingly, since the original input data cannot be accessed, leakage of personal information can be prevented.

  When the detection of personal information and the masking process are completed until the end of the input data (Yes in step S9), the dummy file is deleted (step S10). Subsequently, the write lock on the file database is released (step S11), the file name of the original input data is restored, and the original input data is stored in the file database (step S12).

  Thus, the input data written by the client terminal is registered in the database in a state where the personal information existing in the input data is masked. Accordingly, even when accessed from another client terminal 22 or when data is leaked from the information management server, only the data whose personal information has already been masked is output. Can be prevented.

  The present invention can be applied to a system for an ISP that provides a bulletin board, a blog service, and an uploader on the Internet, and an enterprise file server that holds business data, and has industrial applicability.

It is a block diagram which shows the structure of the whole network to which the file server was connected. It is a functional block diagram which shows the structure of a file server. It is a figure which shows the mode of the change of the input data which a file server processes. It is a flowchart which shows operation | movement of a file server. It is a flowchart which shows operation | movement of a file server. It is a flowchart which shows operation | movement of a file server.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 File server 3 Search server 10 File system 11 Data input reception part 12 Personal information masking module 13 File write lock part 14 Personal information detection part 15 Personal information masking part 16 Data output part 17 Personal information masking dictionary 18 File databases 21 and 22 Client terminal

Claims (11)

An information management device that receives input of predetermined data and stores and manages the input data in a storage device,
Personal information detection means for detecting a personal information part in the input data based on a preset personal information determination criterion;
A data processing means for detecting personal information section data, you pressurized engineering as unrecognizable the personal information section in the personal information detecting means,
Data writing stopping means for stopping writing of the input data to the storage device until the processing by the personal information detecting means and the data processing means for the input data is completed,
The data writing stop means has the same identification information as the identification information for specifying the input data, and generates dummy data having a data content different from that of the input data, and the personal information detection means for the input data And the dummy data is deleted after the processing by the data processing means is completed, and the identification information for specifying the input data is changed to other identification information until the processing by the personal information detection means and the data processing means is completed. Then, after completion of the processing by the personal information detection means and the data processing means, the identification information for specifying the input data is returned to the original, and the input data is stored in the storage device.
An information management apparatus characterized by that. A personal information determination dictionary that stores a preset keyword or character string pattern that is the personal information determination criterion;
The personal information detecting means detects a portion corresponding to a keyword or a character string pattern stored in the personal information determination dictionary in the input data as the personal information portion;
The information management apparatus according to claim 1. The personal information detecting means extracts a noun part in the input data based on a preset noun extraction criterion, makes a search request to a search server accessible via the network, and Detecting the noun part as the personal information part based on a search result from a search server;
The information management apparatus according to claim 1, wherein the information management apparatus is an information management apparatus. The personal information detecting means detects the noun part as the personal information part when the number of search hits by the search result from the search server is equal to or less than a preset threshold among the noun parts.
The information management apparatus according to claim 3. The personal information detecting means registers the personal information part detected based on a search result from the search server in the personal information determination dictionary.
The information management apparatus according to claim 3 or 4, characterized in that An information management device that receives input of predetermined data and stores and manages the input data in a storage device.
Personal information detection means for detecting a personal information part in the input data based on a preset personal information determination criterion;
A data processing means for detecting personal information section data, you pressurized engineering as unrecognizable the personal information section in the personal information detecting means,
Realizing data writing stop means for stopping writing of the input data to the storage device until the processing by the personal information detecting means and the data processing means for the input data is completed, and
The data writing stop means has the same identification information as the identification information for specifying the input data, and generates dummy data having a data content different from that of the input data, and the personal information detection means for the input data And the dummy data is deleted after the processing by the data processing means is completed, and the identification information for specifying the input data is changed to other identification information until the processing by the personal information detection means and the data processing means is completed. Then, after completion of the processing by the personal information detection means and the data processing means, the identification information for specifying the input data is returned to the original, and the input data is stored in the storage device.
A program to make things happen. The personal information detecting means detects, as the personal information part, a part corresponding to a preset keyword or character string pattern that is the personal information determination criterion in the input data;
The program according to claim 6 . The personal information detecting means extracts a noun part in the input data based on a preset noun extraction criterion, makes a search request to a search server accessible via the network, and Detecting the noun part as the personal information part based on a search result from a search server;
8. The program according to claim 6 or 7, wherein: Information management apparatus receives an input of predetermined data, an information management method for managing and storing the input data in the storage device,
The information management device detects a personal information part in the input data based on a personal information determination criterion set in advance,
The information management apparatus, a data detected personal information part in the personal information detecting step, and a data processing step you pressurized engineering as unrecognizable the personal information part,
The information management apparatus, until the processing by the personal information detecting step and the data processing step for the input data is finished, have a, a data write stopping step of stopping the writing to the storage device of the input data,
In the data write stop step, the information management device generates dummy data having the same identification information as the identification information for specifying the input data and having a data content different from that of the input data. The dummy data is deleted after the processing by the personal information detecting step and the data processing step is completed, and the identification information specifying the input data is used until the processing by the personal information detecting step and the data processing step is completed. Change to other identification information, return the identification information specifying the input data after the end of the processing by the personal information detection step and the data processing step, and store the input data in the storage device,
An information management method characterized by that. In the personal information detection step, the information management device detects, in the input data, a portion corresponding to a preset keyword or character string pattern that is the personal information determination criterion as the personal information portion.
The information management method according to claim 9 . In the personal information detection step, the information management device extracts a noun part in the input data based on a preset noun extraction criterion, and a search server that can access the noun part via a network Search request, and detect the noun part as the personal information part based on the search result from the search server,
The information management method according to claim 9 or 10 , characterized in that

Images