JP2007108356A - Personal information concealing device and program for same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a personal information concealing device which not only divides personal information item by item but also processes a name such as a person's name by the unit of a character in a way where kanji (Chinese character) and reading of the kanji are consistent so as to enhance the secrecy of the personal information. <P>SOLUTION: A name constituting input data is divided into respective kanji. Reading of the respective divided kanji is divided on the basis of name reading. One or more of the respective divided kanji and one or more reading of the corresponding kanji are taken out, and the kanji and the reading taken out are combined to newly create a name and name reading. Since the name and the name reading are replaced by the created name and name reading, the kanji and the reading of the kanji of the input data are used and the output data wherein the personal information is concealed can be acquired while maintaining the data property of the input data. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a personal information concealment device that performs concealment processing on personal information.

  With the rapid progress of the IT society, e-government and e-government have been established in public institutions, and e-commerce and customer services have become more sophisticated in private companies, increasing the opportunity to use personal information electronically. ing. Under these circumstances, the Act on the Protection of Personal Information was enforced in April 2005 to harmonize the protection and use of personal information.

Therefore, the handling of personal information within the scope that deviates from the achievement of the specific purpose of use is prohibited in principle, and in principle, the consent of the person is required when providing personal data to a third party.
In order to cope with such a situation, there is a personal information management apparatus disclosed in Japanese Patent Laid-Open No. 2001-265771 as an apparatus that handles personal information.

The personal information management device as the background art registers the personal information provided by the personal information provider terminal device in the personal information management center device through the communication means, and the personal information user terminal device communicates the personal information with the communication means. A personal information management device that is acquired and used via the personal information management center device, wherein the personal information is divided into partial personal information that cannot be associated with a specific individual alone and does not make sense Means for receiving personal information from the personal information provider terminal device; identifier generating means for generating an identifier for the received partial personal information; and storing the generated partial personal information identifier and the partial personal information in correspondence with each other Partial personal information storage means, means for transmitting the partial personal information identifier to a personal information provider terminal device, and partial personal information received from the personal information user terminal device It means for retrieving the partial personal information corresponding to Besshi from the partial personal information storage unit, a configuration in which a means for transmitting said retrieved partial personal information to the personal information provider terminal device.
As described above, according to the personal information management apparatus, partial personal information obtained by dividing the personal information is recorded, and even if it is leaked to the outside, it is difficult to compose the original personal information.
JP 2001-265771 A

  According to the personal information management device of the background art, the personal information is not recorded in the storage means as personal information, but is recorded in the storage means as partial personal information, and the partial personal information is composed as personal information with the consent of the personal information provider. Can be provided.

  However, depending on the method of dividing personal information, the partial personal information itself becomes personal information, and there is a problem that the operation of the personal information management apparatus is very difficult. When personal information consists of multiple items, even if there is no relationship with other items divided by item, the name itself is included in countless information, so the information can be obtained without the consent of the person It will be leaked. For example, when an employee database of a certain company is leaked, even if the relevance of each item itself is lost, information that each individual is an employee of the company is leaked. Similarly, even if the relevance of each item itself is lost, information is leaked regarding information in which all persons have the same value for a certain data item.

  The present invention not only divides the personal information into items, but also processes the names such as names in character units in a form that is consistent with the kanji and the reading of the kanji, and increases the confidentiality of the personal information. An object is to provide an information concealment device.

  (1) The personal information concealment device according to the present invention includes, as input data, personal information including a name that is a character string consisting of at least a kanji character and a name reading that is a character string consisting of a reading for the name. The name is divided into character units, the name reading is divided so as to correspond to the divided character unit kanji, the kanji reading dividing means for associating the divided kanji with the corresponding reading, and the plurality of divided kanji and the corresponding readings. Name composition means for generating a name and a name reading by associating a kanji character string and a corresponding kanji character string with a combination of one or more randomly extracted kanji characters and their corresponding readings. Replace the name in the data with the name generated by the Kanji reading composition means, and the name reading in the input data by the Kanji reading composition means In which and a means for outputting be replaced by the name readings with related to the name form was.

  (2) The personal information concealment device according to the present invention further comprises dictionary data storage means for recording dictionary data composed of kanji in character units and readings for the kanji characters, if necessary, wherein the kanji reading dividing means is separated. Based on the kanji for each character, the dictionary data is referenced to read the reading for the kanji, and when the name reading includes the read reading, the name reading is divided by the corresponding reading and the name is read by the kanji reading dividing means. In the case where the divided name reading corresponding to the divided kanji character is not registered in the dictionary data when the reading is divided, a means for registering the name reading in association with the kanji character in the dictionary data is provided.

(3) The personal information concealment device according to the present invention further includes dictionary data storage means for recording dictionary data consisting of kanji in character units, reading with respect to the kanji, and the number of references, if necessary.
The kanji reading division means reads the reading for the kanji by referring to the dictionary data based on the separated kanji for each character, and if the name reading includes the reading, the name reading is divided by the corresponding reading. The dictionary data includes means for updating the kanji related to the division and the number of references corresponding to the reading for the kanji, and the reading for each kanji constituting the name generated by the name composition means is based on each kanji. The dictionary data is referred to as a reading having the highest number of references, and a name reading is replaced with these readings.

  (4) The personal information concealment device according to the present invention includes, as necessary, master data storage means for recording master data composed of data items corresponding to data items constituting the input data, and names constituting the input data. And means for taking out and replacing the data of the data items other than the name reading at random from the data of the corresponding data items of the master data.

(5) The personal information concealment program according to the present invention uses, as input data, personal information including a name that is a character string consisting of at least a kanji character and a name string consisting of a reading for the name. The input data name is divided into character units, the name reading is divided so as to correspond to the divided character unit kanji, kanji reading dividing means for associating the divided kanji with the corresponding reading, and a plurality of divided kanji characters and Name composition means for generating a name and a name reading by associating a kanji character string composed of kanji characters and a kanji character string consisting of the kanji characters and a kanji character string consisting of the kanji characters. And replace the name in the input data with the name generated by the Kanji reading composition means, Is replaced with a name reading associated with the name generated by the Kanji reading composition means and functions as a means for outputting.
These outlines of the invention do not enumerate the features essential to the present invention, and a sub-combination of these features can also be an invention.

Here, the present invention can be implemented in many different forms. Therefore, it should not be interpreted only by the description of the following embodiment. Also, the same reference numerals are given to the same elements throughout the embodiment.
In the embodiment, the apparatus will be mainly described. However, as is apparent to those skilled in the art, the present invention can also be implemented as a program, system, and method usable on a computer. In addition, the present invention can be implemented in hardware, software, or software and hardware embodiments. The program can be recorded on any computer-readable medium such as a hard disk, CD-ROM, DVD-ROM, optical storage device, or magnetic storage device. Furthermore, the program can be recorded on another computer via a network.

[1. Summary of the Invention]
FIG. 1 is a data flow diagram of a personal information concealment device according to an embodiment of the present invention.
The kanji reading dividing means appropriately divides the data item “name” and the data item “reading” in the input data, and associates one kanji character with the reading of the kanji. When determining reading, dictionary data is referred to as necessary.
-If the new kanji character is targeted, the kanji reading dividing means registers it in the dictionary data together with the reading.
-When the kanji reading has been registered but the reading has not been registered, the kanji reading dividing means registers the new reading in association with the already registered kanji.
-When the kanji and reading are already registered, the kanji reading dividing means updates the reference count of the kanji and reading.
The name composition means generates a new name by replacing one kanji character of one name with one kanji character of another name for the data item “name” in the input data.
The name composition means checks whether the generated name is not the same as the name in the input data, and adopts this name only when it is not the same.
When the name composition means has two characters for the newly generated name, the reading of the second character is determined from the reference number of the corresponding character registered in the dictionary data.
The record generation means replaces the data item of one input data with the data item of one input data of the master data for the data item of the input data. This replacement is performed on all input data for the designated data item.
The record generation means compares the replaced input data with all the input data before replacement, checks whether they are not identical, and adopts this input data only when they are not the same value.

Input data is personal information that is subject to concealment processing.
A data item is an attribute that constitutes input data.
The master data is data for replacing the data items of the input data in the apparatus in advance.
The user can specify which data item of the input data is to be masked using a setting file (layout definition information).

[2. Hardware configuration diagram]
FIG. 2 is a hardware configuration diagram of the personal information concealment device according to the present embodiment. The personal information concealment device is configured, for example, by installing a personal information concealment program in a computer and reading the personal information concealment program into a main memory. The hardware configuration of the computer 10 in this case is a general computer configuration, a main memory such as a CPU (Central Processing Unit) 11 and a DRAM (Dynamic Random Access Memory) 12, and an HD (hard disk) that is an external storage device. ) 13, a display 14 as a display device, a keyboard 15 and a mouse 16 as input devices, a LAN card 17 as an expansion card for connection to a network, a CD-ROM drive 18 and the like.

[3. Data structure diagram]
FIG. 3 is an explanatory diagram of data used in the personal information concealment device according to the present embodiment. 3 (a) is dictionary data, FIG. 3 (b) is regional master data, FIG. 3 (c) is master data about the company, FIG. 3 (d) is master data about the organization of the company, and FIG. Input data, FIG. 3 (f) is setting data. Here, the configuration of each data shown is an example, and a so-called person skilled in the art can appropriately change the design.

The dictionary data has attributes of kanji, reading, and reference count.
Master data must be prepared according to the master item of input data. In the present embodiment, master data related to the region, master data related to the company, and master data related to the organization of the company are prepared and used. The master data relating to the region has attributes of prefecture, prefecture (reading), city, town (town), postal code, and area code. The master data relating to the company has attributes of company name and company name (reading). The master data relating to the organization of the company has the attributes of division name and division name (reading). Note that the master data regarding the surname is generated by the personal information concealment device.

  Input data varies depending on the target personal information. As an example, the input data has attributes of last name, reading (same as last name (reading)), prefecture, city, postal code, card number, company name, department name, and gender.

  The setting data has attributes of data item name, mask item designation, and designation flag. The data item name matches the data item of the input data. When the value “Y” is entered, the designation flag indicates that the data item is to be masked. When the value is not entered, the specification flag indicates that the data item is not to be masked. Show. The mask item designation is the content of the mask process. If the data item name of the master data is included in the mask item specification, mask processing is performed on the data item of the master data. When a character string starting with “$” is included in the mask item specification, a mask process that can specify the process with the character string is performed. “$ N” means that no processing is performed on the corresponding character, and “$ d” means that the corresponding character is masked with a decimal random number.

[4. Explanation using specific data]
4 to 6 are explanatory diagrams using specific data by the personal information concealment device according to the present embodiment.
In the setting file shown in FIG. 3F, the data item names “last name” and “reading” both have the designation flag “Y”, and the data items of the surname and surname (reading) are included in the mask data. Therefore, the Chinese character reading dividing means processes the last name and kana in the input data.

  The dictionary data is searched for “Suzuike”, which is the last name of the first record, and “Suzuike”, which is the reading, using “Suzu” as a key (see FIG. 4A). The Chinese character “bell” is registered in the dictionary data, and the reading “tin” and the reference count “1”, the reading “Rin” and the reference count “1” are read out. Since the first character and the second character of the reading “Suzuike” of the input data match the readings read, the reference count is incremented by 1, and the reference count of the corresponding record of the dictionary data is updated to “2”. Then, the Chinese character “Suzuike” of the input data is divided into “bell” and “pond”, and the reading is also divided into “tin” and “ike” (see FIG. 4B). After the division, the dictionary data is searched using the “pond” of the second character of the input data as a key. Since the Chinese character “pond” is not registered in the dictionary data, the Chinese character “pond”, the reading “Ike” and the reference count “1” are registered.

  The dictionary data is searched for “Yamada” as the last name of the second record and “Yamada” as the reading with “Yama” as a key (see FIG. 4A). The Chinese character “mountain” is registered in the dictionary data, and reading “yama” and reference count “1”, reading “san” and reference count “1” are read out. Since the first and second characters of the input data reading “Yamada” match the readings read, the reference count is incremented by 1 and the reference count of the corresponding record in the dictionary data is updated to “2”. Then, the Chinese character “Yamada” of the input data is divided into “yama” and “da”, and the reading is also divided into “yama” and “da” (see FIG. 4B). After the division, the dictionary data is searched using the “character” of the second character of the input data as a key. In the dictionary data, the Chinese character “da” is registered, but the reading is not registered, so the reading “da” and the reference count “1” are registered for the Chinese character “da”.

  The dictionary data is searched for “ike” as a key for “Ikenaka” which is the last name of the third record and “Ikenaka” which is the reading (see FIG. 4A). The Chinese character “pond” is registered in the dictionary data, and the reading “Ike” and the reference count “2” are read out. Since the first character and the second character of the reading “Ikenaka” of the input data match the readings read, the reference count is incremented by 1, and the reference count of the corresponding record of the dictionary data is updated to “3”. Then, the Chinese character “Ikenaka” of the input data is divided into “Iken” and “Middle”, and the reading is also divided into “Ike” and “Naka” (see FIG. 4B). After the division, the dictionary data is searched by using “middle” of the second character of the input data as a key. Since the Chinese character “middle” is not registered in the dictionary data, the Chinese character “middle”, the reading “Naka”, and the reference count “1” are registered.

  After appropriately dividing the kanji and the reading of the input data, only the second kanji in the input data is generated with random numbers and combined with the first kanji according to the corresponding random numbers. If FIG. 4B is a 3 × 2 array X, a random number between 1 and 3 for X [n] [2] (n = 1, 2, 3) containing the second Kanji character. And the second Chinese character is replaced based on the generated random number (see FIG. 4C).

The first kanji reading is used as it is, and the second kanji is read with the highest number of references using the kanji as a key. X [1] [2] is the Chinese character “middle”, and when the dictionary data is referred to, it is read “Naka” (see FIG. 4D). X [2] [2] is the Chinese character “pond”, and when the dictionary data is referenced, it reads “Ike”. X [3] [2] is the Chinese character “Ta” and is read as “Ta” when the dictionary data is referenced. Here, only the Chinese character “da” is read from “da” and becomes “ta”.
By these processes, master data regarding the surname is generated.
As shown in FIG. 5, each master data is stored in an array that matches the number of data items. Then, the array element number of each master data array is determined by a random number.

The data item “Suzuike” of the last name of the first record of the input data is replaced with the last name element “Ikeda” stored in the array number “0” of the master data relating to the last name (see FIG. 6A). The reading data item “Suzuike” in the first record of the input data is replaced with the reading element “Ikeita” stored in the array number “0” of the master data regarding the surname (see FIG. 6B). . The prefecture “Tokyo” in the first record of the input data is replaced with the element “Tokushima” of the prefecture stored in the array number “0” of the master data relating to the region (see FIG. 6C). The municipality “Ota-ku Iwata” in the first record of the input data is replaced with the element “Tokushima-shi Minamiokisu” stored in the array number “0” of the master data relating to the region (FIG. 6 ( d)). The zip code “114-0959” of the first record of the input data is replaced with the zip code element “770-0874” stored in the array number “0” of the master data relating to the region (see FIG. 6E). ). The card number “1246-3371” of the first record of the input data is the first to fourth according to the mask item “$ d $ d $ d $ d- $ n $ n $ n $ n” of the card number of the setting file. The numbers up to are obtained by random numbers, and the numbers from the fifth to the eighth are unchanged and become “5761-3371”, which are replaced with these numbers (see FIG. 6F). The company name “XX sightseeing” in the first record of the input data is replaced with the element “D department store” of the company name stored in the array number “0” of the master data relating to the company (see FIG. 6G). . The department name “management department” of the first record of the input data is replaced with the element “distribution department” of the department name stored in the array number “0” of the master data related to the organization of the company (see FIG. 6H). ). The gender “male” in the first record of the input data is not replaced because the gender designation flag in the setting file is not “Y” (see FIG. 6I).
Thereafter, the second record and the third record of the input data are similarly replaced. Eventually, the input data of FIG. 6J becomes the output data of FIG. 6K by the information concealment process.

[5. flowchart]
The use operation of the personal information concealment device according to the present embodiment will be described with reference to the drawings.
FIG. 7 is a kana-kana dividing operation flowchart of the personal information concealment device according to the present embodiment. First, the processor 11 acquires the last name and reading of the target input data (step 101). The processor 11 searches and determines whether or not the first Chinese character of the last name of the input data is registered in the dictionary data (step 111). If it is determined that it is registered in the dictionary data, the processor 11 determines whether or not the reading of the input data matches the reading of the dictionary data (step 121). If it is determined that the reading of the input data matches the reading of the dictionary data, the corresponding kanji in the dictionary data and the number of times of reading are updated (step 122). Next, the processor 11 determines whether or not there is a second Chinese character in the last name of the input data (step 131). If it is determined that there is no second kanji, the process ends. If it is determined that there is a second character, the last name and the reading of the input data are divided into the first character, the first character, and the second and second characters (step 132). . The processor determines whether the second kanji and the second character reading are registered in the dictionary data (step 141). If kanji and reading are registered, the processor 11 updates the reference count of the dictionary data (step 142), and the process ends. If only kanji is registered, the second character reading is registered in association with the kanji and the reference count is set to “1”. If not even kanji is registered, the second kanji and the second character are registered. The reading is registered and the reference count is set to “1” (step 143).

  If it is determined in step 111 that it is not registered in the dictionary data, or if it is determined in step 121 that the reading does not match the reading of the dictionary data, the processor 11 adds the second character to the last name of the input data. It is determined whether or not there is a Chinese character (step 151). If it is determined that there is a second character, the processor 11 determines whether the second character of the last name of the input data is in the dictionary data (step 161). If it is determined that the second character is not in the dictionary data, the input data is forcibly divided into two (step 162), and the process ends. Forcibly dividing into two is, for example, dividing the first kanji and the second kanji, and dividing the reading into the first reading and the subsequent reading, and the first kanji and 1 The reading of the character is made to correspond to the reading of the second character and the reading of the first character. In this case, it is not always possible to determine whether kanji and reading are correct, so it is desirable not to register the dictionary data. If it is determined in step 161 that the second character is in the dictionary data, the processor 11 determines whether the reading in the dictionary data corresponding to the second character matches the reading in the input data. Is determined (step 171). If it is determined that the readings do not match, the process proceeds to step 162. If it is determined in step 171 that the readings match, the processor 11 updates the reference count of the dictionary data for the second kanji and the second character reading (step 172). The processor 11 divides the last name and reading of the input data based on the second kanji and the second reading (step 181). Next, when only the kanji is registered, the first character reading is registered in association with the kanji and the reference count is set to “1”. When even the kanji is not registered, the first kanji and The reading of the first character is registered and the reference count is set to “1” (step 191).

In FIG. 7, “◯” indicates a kanji in which kanji is registered in the dictionary data and the reading is also registered, “Δ” indicates a kanji in which kanji is registered in the dictionary data and the reading is not registered, and “×” "Indicates a kanji that is not even registered in the dictionary data. Therefore, XX means two kanji characters, and both kanji indicate that kanji and readings are registered in the dictionary data. FIG. 7 shows the position of the flowchart that is finally reached when each symbol is a processing target.
The kanji kana dividing operation shown in FIG. 7 described above is performed for each surname and kana of the input data.

  FIG. 8 is a composition operation flowchart of the personal information concealment device according to the present embodiment. First, the processor 11 generates random numbers for all surnames and readings of the input data to change the combination of the first character and the second character of the surname (step 201). For the last name of interest, the second kanji for the first kanji is determined using random numbers from the second kanji for all surnames in the input data. In addition, with respect to the last name to be processed, the first kanji for the second kanji may be determined using random numbers from the first kanji for all surnames in the input data. Also, for the last name of interest, the first kanji is determined using random numbers from the first kanji of all surnames in the input data, and the second kanji is the two characters of all surnames in the input data. It may be determined using a random number from the kanji of the eye.

  Next, it is determined whether or not the new kanji composed by the processor 11 is not the same as the last name of the input data (step 202). If it is determined that they are the same, the process returns to step 201. If it is determined that they are not identical, the kanji readings constituting the newly formed surname are searched for dictionary data based on the target kanji and replaced with readings with the highest number of references (step 203). If no kanji is registered in the dictionary data, the current reading is used. Moreover, readings may be replaced only for kanji determined using random numbers, or readings may be replaced using dictionary data for all kanji.

  The composition operation of FIG. 8 described above is performed for each surname and kana of the input data. Here, the last name of the data item of the input data can also be regarded as being composed of a first kanji group of a plurality of surnames and a second kanji group of a plurality of last names. As described above [4. In the description using specific data], a new random number is used from a group of second kanji characters of a plurality of surnames excluding the second kanji character already determined using random numbers in the process of step 201. The second character is determined. A Chinese character may be newly determined using a random number from a group of Chinese characters excluding the determined Chinese character, but a new Chinese character can also be determined using a random number without removing the determined Chinese character. .

  FIG. 9 is a record generation operation flowchart of the personal information concealment device according to the present embodiment. First, the processor 11 reads master data into the main memory 12 (step 301). The processor 11 rearranges the master data array using random numbers. Specifically, rearrangement can be performed by determining the array element number (array subscript) of the array storing each master data by a random number and rearranging the array. The processor 11 initializes the replacement candidate counter i (step 321). For example, i = 1. A loop is entered (step 331), and the processor 11 processes one record from the input data (step 341). Further, a loop is entered (step 351), and the processor 11 refers to the setting data to determine whether or not the target data item is a mask target (step 361). If it is determined that it is a mask target, the processor 11 determines whether or not the target data item is character data designation (step 371). When it is determined that the character data is not designated, the data of the target data item is replaced with the data of the corresponding data item in the master data associated with the setting data (step 371). When the processor 11 has the next data item, the process returns to step 351 for the data item (step 352).

  If it is determined in step 371 that the target data item is character data designation, the process proceeds to a predefined process numerical data generation process (step 500). This predefined process will be described later. After this processing, the processor 11 replaces the data of the target data item with the generated data (step 391), and proceeds to step 352.

  Next, it is determined whether or not the input data replaced by the processor 11 is the same (step 401). If it is determined that they are the same, the replaced input data is excluded, and the processor 11 decrements the replacement candidate counter i (step 402). Here, the comparison between the replaced input data and the input data may be a completely matched configuration or a partially matched configuration. For example, when the data items such as key candidates uniquely identify the data items, it is desirable to proceed to step 402.

  Then, it is determined whether or not the number of input data replaced by the processor 11 so far has been generated (step 411). If it is determined that it has been generated, the process ends. If it is determined that it has not been generated, the replacement candidate counter i is incremented, and if there is a record of the next input data, the process returns to step 331 for that record (step 332). If there is no next record, the process goes out of the loop from step 311 to step 332, and it is determined whether or not the number of input data replaced by the processor 11 up to the present time has been generated (step 421). If it is determined that it has been generated, the process ends. If it is determined that it has not been generated, the process returns to step 311.

  The predefined process numerical data generation process will be described next. FIG. 10 is a flowchart of the numerical value generation operation of the personal information concealment device according to the present embodiment. First, the processor 11 acquires the corresponding data of the target data item of the input data (step 501). The processor 11 acquires the designated character string (numerical value designation) of the mask item of the setting data (step 511). A loop is entered (step 521), and the processor 11 acquires the next two characters of the designated character string (step 531). It is determined whether or not the character acquired by the processor 11 is “$ d” (step 541). If it is determined that it is $ d, a number from 0 to 9 is output as a random number (step 542). If it is determined that it is not “$ d”, it is determined whether or not the character acquired by the processor 11 is “$ n” (step 551). If it is determined that it is “$ n”, the data of the target data item of the input data at the same position as “$ n” is output (step 552). If it is determined that it is not “$ n”, the mask item designation character is output as it is (step 553). If there is no next designated character, the loop of step 521 to step 522 is exited and the process is terminated. If there is a next designated character, the process returns to step 521.

[6. Effects of this embodiment]
As described above, in the present invention, the Chinese characters constituting the surname are separated by using the surname and reading existing in the input data, and the reading is appropriately separated along with the separation of the kanji, and a new surname is newly formed from the separated kanji. Since the reading corresponding to the surname is given, the kanji and reading that exist in the input data are used as they are, and the output is retained while maintaining the data characteristics of the input data while concealing personal information. Data can be obtained. Moreover, since the dictionary data is updated along with the separation of kanji and readings, readings can be appropriately separated and appropriate readings can be given to the composed kanji. The data items of the input data other than the surname and reading can also be replaced at random as necessary, and the test data desired by the user can be obtained.

(Other embodiments)
[When Name (Last Name) Is More Than 3 Characters] In the first embodiment, a case where the last name is 1 character and 2 characters is explained, and many surnames can be processed. In the case of three characters, the first two characters are processed as a name consisting of two characters, and the first character and its reading are obtained. Then, the reading of the first character (reading of the first character) and the reading of the second character (reading of the second character) are inevitably determined. Can be determined. Here, since the first kanji and its reading cannot always be determined, the third kanji reading is reversed as the third kanji dictionary. It is also possible to adopt a configuration in which the determination is made using data, and then the reading of the first kanji is determined.

Also, even in the case of the third Chinese character, the reference count is updated according to the reference, and if it is not registered, registration is also performed.
Even in the case of a three-character kanji surname, the step 201 can be used to convert a three-character kanji last name into another three-character kanji last name. However, since a surname of 3 characters or more can be used as information that can identify an individual, it is desirable to exclude the 3rd and subsequent kanji characters and make up the last name with 2 kanji characters. . However, the third and subsequent kanji can also be used as the kanji selected when determining the surname.

  In this way, by applying the present invention to a name having three or more characters, the entire first and last name can be concealed. Further, when the first and last names are distinguished by a character such as a space, a new last name that seems to be a last name and a new name that is more likely to be a first name can be generated by processing each name.

  [Concealment Process for Names Containing Hiragana, Katakana, etc. Other than Kanji] In the first embodiment, the case where the surname consists of kanji has been described, but in the case where the surname consists of characters other than kanji, such as hiragana, katakana, etc. The reading can be divided without referring to the dictionary data and can be handled easily. However, it is desirable to exclude characters other than kanji such as hiragana and katakana as characters to be selected when determining the surname.

There are few surnames that contain hiragana and katakana, but there are many individuals whose names contain hiragana and katakana. There are many companies with company names.
Although the present invention has been described with the above embodiments, the technical scope of the present invention is not limited to the scope described in the embodiments, and various modifications or improvements can be added to these embodiments. . And embodiment which added such a change or improvement is also contained in the technical scope of the present invention. This is apparent from the claims and the means for solving the problems.

[Appendix] The following appendices are further disclosed with respect to the embodiment.
(Supplementary note 1) Personal information including at least a name that is a character string consisting of kanji, including name reading that is a character string consisting of a reading for the name is used as input data, and the name of the input data is divided in character units. The name reading is divided so that it corresponds to the character-by-character kanji, the kanji reading dividing means for associating the divided kanji with the corresponding reading, the kanji divided from the divided kanji and the corresponding readings, and the corresponding readings Name composition means for generating one name and name reading by associating a character string consisting of kanji and a reading corresponding to the character string consisting of kanji characters by combining one or more at random, and kanji reading composition means for the name in the input data The name reading in the input data and the name reading associated with the name generated by the kanji reading composition means A personal information concealment device comprising:

  Therefore, the name constituting the input data is divided for each kanji, the divided kanji readings are divided from the name reading, and one or more are extracted from each divided kanji and the reading corresponding to the kanji, and the extracted kanji and Since the name and name reading are newly generated by combining the readings, and the name and name reading of the input data are replaced with the generated name and name reading, the kanji and reading of the input data are used, and the data characteristics of the input data Output data in which personal information is concealed can be obtained.

The name includes the name, the name of the group / organization / building, the name and name, and the name and name of the thing.
In particular, it is possible to more securely conceal personal information by checking whether or not the generated name is duplicated with the name in the input data, and in the case of duplication, the new name is not adopted.

    (Additional remark 2) It further has a dictionary data storage means which records the dictionary data which consists of the kanji of a character unit and the reading with respect to the said kanji, The said kanji reading division | segmentation means refers to dictionary data based on the separated kanji of a character unit, Reads kanji readings, if name readings include readings that are read, divides name readings by corresponding readings, and supports divided kanji when name readings are divided by the kanji reading dividing means The personal information concealment device according to appendix 1, further comprising means for registering the name reading in the dictionary data in association with the kanji when the divided name reading to be performed is not registered in the dictionary data.

  Therefore, when dividing a name reading, it is divided by referring to the dictionary data, and when the reading corresponding to the divided kanji is not registered in the dictionary data, the reading is newly registered in association with the kanji, so the personal information is stored. Every time the concealment process is executed, the dictionary data is updated, a new reading is registered, and the name reading can be more appropriately divided.

    (Additional remark 3) It further has a dictionary data storage means which records the dictionary data which consists of the kanji of a character unit, the reading with respect to the said kanji, and the frequency | count of reference, and the said kanji reading division | segmentation means refers to dictionary data based on the separated kanji of the character unit The name reading is divided by the corresponding reading when the name reading includes the reading, and the dictionary data corresponds to the kanji related to the division and the reading for the kanji. Means for updating the number of times of reference to be performed, and the reading of each kanji that constitutes the name generated by the name composition means is referred to the dictionary data based on each kanji and is set to the reading with the highest number of references, The personal information concealment device according to appendix 1, comprising means for replacing by reading.

  Therefore, when dividing a name reading, it is divided by referring to the dictionary data, the kanji used at the time of division and the reference count corresponding to the corresponding reading are updated, and the generated name reading is read for each reading of the kanji constituting the name. Since the dictionary data is replaced with the reading with the highest number of references among the readings for the kanji, the name reading is composed of the most frequently used readings, and it is possible to improve that the name reading for the name becomes unnatural.

    (Supplementary Note 4) Master data storage means for recording master data composed of data of data items corresponding to data items constituting input data, and data of data items other than name and name reading constituting input data The personal information concealment device according to appendix 1, further comprising means for randomly extracting and replacing data of corresponding data items.

  Therefore, since the data of the corresponding data item of the master data is randomly extracted and replaced with the data of the input data, the personal information is replaced with the data of the master data of the corresponding data item, and the personal information can be kept secret.

    (Appendix 5) A plurality of data items having a relationship between data items in the master data are recorded as a group of data items in the master data storage means, and the data items to be rearranged in the master data constitute the data item group In such a case, the personal information concealment device according to appendix 4, wherein the data item group is randomly extracted and the input data is collectively replaced with the data item group data.

  Therefore, the corresponding data item of the master data is randomly extracted in units of data item groups and replaced with the data of the input data. If there is a relationship between the data items, the data of the input data is retained while maintaining the relationship. It can be replaced with data. Specifically, if the address consists of data items for prefectures, cities, and towns, the data for the data items for the prefectures, cities, and towns are retrieved to the prefectures, cities, and towns, respectively. Instead, the prefectures, cities, towns and villages are extracted in a lump, and the input data can be input as multiple data items that are not consistent between data items, such as “Fukuoka Prefecture”, “Shibuya Ward”, and “2 Chome”. Records are not replaced.

(Appendix 6)
At least a name that is a character string consisting of kanji characters, personal information including a name reading that is a character string consisting of readings for the name is used as input data, the name of the input data is divided in character units, and the divided character unit A kanji reading dividing means for dividing a name reading so as to correspond to a kanji and associating the divided kanji with the corresponding reading, and a kanji divided from a plurality of divided kanji and corresponding readings and a corresponding reading at random 1 Name composition means for generating a name and a name reading by associating a character string composed of kanji and a reading corresponding to the character string composed of the kanji characters by combining them, and a name generated by the kanji reading composition means, respectively, in the input data And replace the name reading in the input data with the name reading associated with the name generated by the Kanji reading composition means. Personal information confidential program for functioning as means for outputting Te.

  The output data generated by the personal information concealment device of the present invention can be used as test data that is very close to actual data in a test process for creating a program. That is, test data having data characteristics of actual data can be used. Therefore, the test can be performed in a form close to a real environment by using the test data output after the data used in the old system at the system construction destination is concealed by the personal information concealment device.

Traditionally, programmers, system engineers, and quality assurance departments have created not only test programs but also test data according to the program modules. Often, test data differed from actual data.
Personal information can also be used as test data, but it cannot be used easily because it requires a plurality of procedures for managing personal information.

It is a data flow figure of the personal information concealment device concerning the embodiment of the present invention. It is a hardware block diagram of the personal information concealment apparatus which concerns on embodiment of this invention. It is data explanatory drawing used with the personal information concealment apparatus which concerns on embodiment of this invention. It is explanatory drawing using the concrete data by the personal information concealment apparatus which concerns on embodiment of this invention. It is explanatory drawing using the concrete data by the personal information concealment apparatus which concerns on embodiment of this invention. It is explanatory drawing using the concrete data by the personal information concealment apparatus which concerns on embodiment of this invention. It is a Chinese character kana division | segmentation operation | movement flowchart of the personal information concealment apparatus which concerns on embodiment of this invention. It is a composition operation | movement flowchart of the personal information concealment apparatus which concerns on embodiment of this invention. It is a record production | generation operation | movement flowchart of the personal information concealment apparatus which concerns on embodiment of this invention. It is a numerical value generation operation | movement flowchart of the personal information concealment apparatus which concerns on embodiment of this invention.

Explanation of symbols

10 Computer 11 CPU (Processor)
12 DRAM
13 HD
14 Display 15 Keyboard 16 Mouse 17 LAN card 18 CD-ROM drive

Claims (5)

At least a name that is a character string consisting of kanji characters, personal information including a name reading that is a character string consisting of readings for the name is used as input data, the name of the input data is divided in character units, and the divided character unit A kanji reading dividing means for dividing the name reading so as to correspond to the kanji and associating the divided kanji with the corresponding reading;
Name and name by associating one or more randomly extracted kanji characters and corresponding readings from a plurality of divided kanji characters and corresponding readings and associating them with character strings consisting of kanji characters and readings corresponding to character strings consisting of the kanji characters. Name composition means to generate as a reading;
Personal information comprising means for replacing the name in the input data with the name generated by the kanji reading composition means, and replacing the name reading in the input data with the name reading associated with the name generated by the kanji reading composition means Concealment device. It further comprises dictionary data storage means for recording dictionary data consisting of kanji in character units and readings for the kanji,
The kanji reading division means reads the reading for the kanji by referring to the dictionary data based on the separated kanji for each character, and if the name reading includes the reading, the name reading is divided by the corresponding reading. And
Means for registering the name reading in the dictionary data in association with the kanji when the name reading corresponding to the divided kanji is not registered in the dictionary data when the name reading is divided by the kanji reading dividing means. The personal information concealment device according to claim 1. It further comprises dictionary data storage means for recording dictionary data consisting of kanji in character units, reading for the kanji, and the number of references,
The kanji reading division means reads the reading for the kanji by referring to the dictionary data based on the separated kanji for each character, and if the name reading includes the reading, the name reading is divided by the corresponding reading. And
Means for updating the number of references corresponding to the kanji for the division of the dictionary data and the reading of the kanji,
A means for replacing the reading of each kanji that constitutes the name generated by the name composition means with the highest number of references by referring to the dictionary data based on each kanji and replacing the name reading with these readings. The personal information concealment device described in 1. Master data storage means for recording master data composed of data of data items corresponding to data items constituting input data;
The personal information concealment device according to claim 1, further comprising: means for randomly extracting and replacing data of data items corresponding to master data for data items other than names and name readings constituting input data. Computer
At least a name that is a character string consisting of kanji characters, personal information including a name reading that is a character string consisting of a reading for the name is used as input data, the name of the input data is divided in character units, and the divided character unit A kanji reading dividing means for dividing the name reading so as to correspond to the kanji and associating the divided kanji with the corresponding reading;
Name and name by associating one or more randomly extracted kanji characters and corresponding readings from a plurality of divided kanji characters and corresponding readings and associating the kanji character strings and the corresponding kanji character string readings. Name composition means to generate as a reading;
To replace the name in the input data with the name generated by the Kanji reading composition means, and to function as a means to output the name reading in the input data by replacing it with the name reading associated with the name generated by the Kanji reading composition means Personal information concealment program.

Images