Anonymization system and anonymization method

The anonymization system addresses the challenge of ensuring data validity by using a multi-device approach to generate and verify hash values, ensuring the integrity of anonymized data and preventing fraudulent use.

JP7596248B2Active Publication Date: 2024-12-09HITACHI LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2021170407
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-10-18
Publication Date
2024-12-09
Estimated Expiration
2041-10-18

AI Technical Summary

Technical Problem

The unlinkable anonymization method in existing technologies cannot guarantee the validity of anonymization for entire data sets, leading to potential fraudulent use of anonymized data and inappropriate measures or services based on it.

Method used

An anonymization system comprising a first device, a second device, and a third device that generate and verify hash values to ensure the validity of anonymization processing, using a first hash value generation, anonymization processing, and verification processes to confirm the integrity of anonymized data.

Benefits of technology

Ensures the validity of anonymization processing even after data has been anonymized, preventing the provision of fraudulent data and ensuring the legitimacy of services based on anonymized data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007596248000001
    Figure 0007596248000001
  • Figure 0007596248000002
    Figure 0007596248000002
  • Figure 0007596248000003
    Figure 0007596248000003
Patent Text Reader

Abstract

To verify the correctness of anonymization processing on a data group even after the anonymization processing is performed on pieces of data in the data group.SOLUTION: A first apparatus creates a second hash value of a plurality of data groups different in attribute based on a first hash value of pieces of data in the plurality of data groups. A second apparatus creates a first hash value of pieces of data in a data group with a specific attribute in response to an anonymization request from a third apparatus. As a result of transmitting the anonymization request for the data group with the specific attribute to the second apparatus, the third apparatus acquires the first hash value of the pieces of data in the data group with the specific attribute, a data group with a residual attribute excluding the data group with the specific attribute, and the second hash value, creates a third hash value based on the first hash value of the pieces of data in the data group with the residual attribute and the first hash value of the pieces of data in the data group with the specific attribute, and verifies the correctness of anonymization processing by using the second hash value and the third hash value.SELECTED DRAWING: Figure 10
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an anonymization system and an anonymization method for performing anonymization. [Background technology]

[0002] The Act on the Protection of Personal Information has led to the increased use of anonymously processed information, in which descriptions that could identify specific individuals have been deleted or replaced. The Act to Partially Amend the Act on the Protection of Personal Information, etc. (commonly known as the Amended Personal Information Act of 2020), which is scheduled to be fully implemented in April 2022, will introduce a new category of "pseudonymized information" that will ease some of the obligations under conditions such as limiting the use to internal analysis, and is expected to lead to further advances in the use of data.

[0003] In data utilization, ensuring that the data has not been improperly altered (data validity) is important in order to guarantee the validity of the results of data utilization. Patent Document 1 describes a technology for verifying the validity of anonymization processing.

[0004] The unlinkable anonymization method of Patent Document 1 creates an anonymized number by using a one-way function such as a hash value calculation to combine an ID number or information that can identify an individual, or that ID information and a key symbol used at the time of anonymization, or related information such as a specimen number that cannot identify an individual by itself, deletes the combination table of anonymized numbers and personal information through unlinkable anonymization, prevents inference of the original individual or specimen number from the anonymized number by using the one-way function, and limits access to the anonymized information to only the owner or authorized person of the information who knows the anonymization key information. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] International Publication No. 2008 / 069011 Summary of the Invention [Problem to be solved by the invention]

[0006] However, the unlinkable anonymization method of Patent Document 1 cannot guarantee the validity of the anonymization for the entire data. Generally, digital signature technology is used as a technology for guaranteeing the validity of data. However, when a digital signature is simply applied, if anonymization processing such as data deletion or generalization is performed, the validity of the data cannot be verified. If fraudulent anonymized data is used, the knowledge obtained from the data will also be fraudulent, and there is a risk that measures and services based on the data will be inappropriate.

[0007] The present invention aims to make it possible to verify the legitimacy of anonymization processing performed on a data group even after the data in the data group has been anonymized. [Means for solving the problem]

[0008] An anonymization system according to one aspect of the invention disclosed in the present application includes a first device, a second device capable of communicating with the first device, and a third device capable of communicating with the second device, the anonymization system including: a first generation process in which the first device generates a second hash value for a plurality of data groups having different attributes based on a first hash value of each piece of data in the plurality of data groups; a first acquisition process in which the second device acquires the plurality of data groups and the second hash value generated by the first generation process; an anonymization process in which the second device generates a first hash value for each piece of data in a data group having a specific attribute among the plurality of data groups in response to an anonymization request from the third device; and an anonymization process in which the third device, as a result of transmitting an anonymization request for the data group of the specific attribute to the second device, generates a first hash value for each piece of data in the data group of the specific attribute from the second device and acquires a second hash value for the data group of the specific attribute among the plurality of data groups. Excludinga second acquisition process for acquiring a data group of a remaining attribute and the second hash value; a second generation process for generating a first hash value for each piece of the data group of the remaining attribute acquired by the second acquisition process, and generating a third hash value based on the first hash value for each piece of the data group of the remaining attribute and the first hash value for each piece of the data group of the specific attribute acquired by the second acquisition process; and a verification process for verifying the validity of the anonymization process by the third device by determining whether the second hash value acquired by the second acquisition process and the third hash value generated by the second generation process match. in the first generation process, the first device generates a first hash value of the data in accordance with a processing rule corresponding to the attribute, and generates a second hash value of the multiple data groups based on the first hash value of the data; in the first acquisition process, the second device acquires the processing rule from the first device; in the anonymization process, the second device generates a first hash value of each data of the data group having the specific attribute in accordance with the processing rule; in the second acquisition process, the third device acquires the processing rule from the second device; and in the second generation process, the third device generates a first hash value of each data of the data group having the remaining attributes in accordance with the processing rule. An anonymization system according to another aspect of the invention disclosed in the present application includes a first device, a second device capable of communicating with the first device, and a third device capable of communicating with the second device, the anonymization system including a first generation process in which the first device generates a second hash value for a plurality of data groups having different attributes based on a first hash value for each piece of data in the plurality of data groups; a first acquisition process in which the second device acquires the plurality of data groups and the second hash value generated by the first generation process; an anonymization process in which the second device generates a first hash value for each piece of data in a data group having a specific attribute among the plurality of data groups in response to an anonymization request from the third device; a second acquisition process in which the third device acquires, as a result of transmitting an anonymization request for the data group having the specific attribute to the second device, the first hash value for each piece of data in the data group having the specific attribute from the second device, a data group of remaining attributes among the plurality of data groups excluding the data group of the specific attribute, and the second hash value; a second generation process for generating a third hash value based on a first hash value of each data of the data group with the specific attribute and the first hash value of each data of the data group with the specific attribute acquired by the second acquisition process; and a verification process for verifying the validity of the anonymization process by the third device by determining whether the second hash value acquired by the second acquisition process matches the third hash value generated by the second generation process. In the anonymization process, the second device generates a first hash value of a portion of the data of the specific attribute for which anonymization is requested, and in the second acquisition process, the third device acquires, as a result of transmitting the anonymization request for the portion of the data of the specific attribute to the second device, the first hash value of the portion of the data of the specific attribute from the second device, a remaining data group excluding the portion of the data of the specific attribute from the multiple data groups, and the second hash value. In the second generation process, the third device generates a first hash value of each data of the remaining data group acquired by the second acquisition process, and the first hash value of each data of the data group with the remaining attribute.and generating the third hash value based on a first hash value of the part of data of the specific attribute acquired by the second acquisition process. Effect of the Invention

[0009] According to a representative embodiment of the present invention, even after the data of a data group has been anonymized, it is possible to verify the validity of the anonymization process for the data group. Problems, configurations, and effects other than those described above will become clear from the description of the following examples. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is an explanatory diagram illustrating an example of a system configuration of an anonymization system. [Diagram 2] FIG. 2 is a block diagram showing an example of the hardware configuration of a computer. [Diagram 3] FIG. 3 is a block diagram illustrating an example of a functional configuration of the signature generator terminal. [Figure 4] FIG. 4 is an explanatory diagram illustrating an example of the sensitive data table. [Diagram 5] FIG. 5 is an explanatory diagram illustrating an example of a random number data table. [Figure 6] FIG. 6 is an explanatory diagram illustrating an example of the generalization rule table. [Figure 7] FIG. 7 is a block diagram illustrating an example of a functional configuration of the anonymized data providing server. [Figure 8]FIG. 8 is an explanatory diagram illustrating an example of the anonymized data table. [Figure 9] FIG. 9 is a block diagram illustrating an example of a functional configuration of the anonymized data providing server. [Figure 10] FIG. 10 is a sequence diagram of the anonymization system. [Figure 11] FIG. 11 is an explanatory diagram showing an example of the processing rule setting screen. [Figure 12] FIG. 12 is an explanatory diagram showing an example of the anonymization request setting screen. [Figure 13] FIG. 13 is a flowchart illustrating a detailed example of the process steps of the signature generation process (step S1003) illustrated in FIG. [Figure 14] FIG. 14 is a flowchart illustrating a detailed example of the cell hash value generation process (step S1301) illustrated in FIG. [Figure 15] FIG. 15 is a flowchart illustrating a detailed example of the process of generating a cell hash value (step S1407) illustrated in FIG. [Figure 16] FIG. 16 is a flowchart illustrating a detailed example of a procedure for the hash value aggregating process. [Figure 17] FIG. 17 is a flowchart illustrating a detailed example of the procedure of the anonymization process (step S1010) illustrated in FIG. [Figure 18] FIG. 18 is a flowchart illustrating a detailed example of the procedure of the anonymization process (step S1704) illustrated in FIG. [Figure 19] FIG. 19 is a flowchart illustrating a detailed example of the procedure of the anonymization process (step S1808) of the attribute value when the processing request shown in FIG. 18 is "generalize." [Figure 20] FIG. 20 is a flowchart illustrating a detailed example of the signature verification process (step S1012) illustrated in FIG. [Figure 21] FIG. 21 is a flowchart illustrating a detailed example of the anonymized global hash value generation process (step S2001) illustrated in FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] <Anonymization system> FIG. 1 is an explanatory diagram showing an example of the system configuration of an anonymization system. The anonymization system 100 verifies that the provided information has not been improperly altered when the information is anonymized and provided to prevent leakage of personal information or sensitive information. The anonymization system 100 is a system that enables a data owner who holds data including sensitive information (hereinafter, sensitive data) to anonymize the information and provide it to a data user. The anonymization system 100 has a signature generator terminal 101, an anonymized data providing server 102, and an anonymized data user terminal 103.

[0012] The signature generator terminal 101, the anonymized data providing server 102, and the anonymized data user terminal 103 are connected so as to be able to transmit and receive information via a network 104. Possible types of the network 104 include the Internet, a Wide Area Network (WAN), or a Local Area Network (LAN). The connection method of the network 104 may be either wired or wireless.

[0013] The signature generator terminal 101 sets an anonymization method (hereinafter referred to as an attribute processing rule) that a signature generator (e.g., a government agency) which is a data holder permits for each attribute of sensitive data. The signature generator terminal 101 generates a signature for the sensitive data so that the validity of the anonymized data that has been anonymized according to the attribute processing rule can be verified.

[0014] The anonymized data providing server 102 provides a signature value of sensitive data, and generates and provides anonymized data from the sensitive data entrusted by the signature generator in accordance with a request from an anonymized data user. The anonymized data user terminal 103 is a terminal used by an anonymized data user, and verifies the validity of the anonymized data.

[0015] <Example of Hardware Configuration of Computers (Signature Generator Terminal 101, Anonymized Data Providing Server 102, and Anonymized Data User Terminal 103)> FIG. 2 is a block diagram showing an example of a hardware configuration of a computer. The computer 200 includes a processor 201, a storage device 202, an input device 203, an output device 204, and a communication interface (communication IF) 205. The processor 201, the storage device 202, the input device 203, the output device 204, and the communication IF 205 are connected by a bus 206. The processor 201 controls the computer 200. The storage device 202 serves as a working area for the processor 201. The storage device 202 is a non-transient or temporary recording medium that stores various programs and data. Examples of the storage device 202 include a ROM (Read Only Memory), a RAM (Random Access Memory), a HDD (Hard Disk Drive), and a flash memory. The input device 203 inputs data. Examples of the input device 203 include a keyboard, a mouse, a touch panel, a numeric keypad, a scanner, a microphone, and a sensor. The output device 204 outputs data. The output device 204 may be, for example, a display, a printer, or a speaker. The communication IF 205 is connected to the network 104 and transmits and receives data.

[0016] <Example of Functional Configuration of Signature Generator Terminal 101> 3 is a block diagram showing an example of a functional configuration of the signature generator terminal 101. The signature generator terminal 101 has a random number data generation unit 301, a signature generation unit 302, a hash value generation unit 303, a hash value aggregation unit 304, a sensitive data table 305, a random number data table 306, a generalization rule table 307, attribute processing rules 308, a signature value table 309, and a signature key 310.

[0017] The random number data generation unit 301, the signature generation unit 302, the hash value generation unit 303, and the hash value aggregation unit 304 are specifically realized, for example, by making the processor 201 execute a program stored in the storage device 202 shown in Fig. 2. In addition, the sensitive data table 305, the random number data table 306, the generalization rule table 307, the attribute processing rule 308, the signature value table 309, and the signature key 310 are stored in the storage device 202.

[0018] The random number data generation unit 301 generates random numbers to be added to enhance the security of hash values. The signature generation unit 302 generates a signature value for sensitive data. The hash value generation unit 303 generates hash values ​​using a one-way function or the like. The hash value aggregation unit 304 aggregates multiple hash values ​​to generate one hash value.

[0019] The sensitive data table 305 stores sensitive data such as personal information. The random number data table 306 stores random numbers that are assigned to increase the security of hash values. Details of the random number data table 306 will be described later with reference to FIG. 5. The generalization rule table 307 stores generalization rules for generalizing sensitive data. Details of the generalization rule table 307 will be described later with reference to FIG. 6.

[0020] The attribute processing rules 308 store processing rules for each attribute of sensitive data. The attribute processing rules 308 include, for example, "pseudonymization," "deletion," "generalization," and "no processing allowed." "Pseudonymization" is a processing rule that removes attributes indicating personal information such as a name from the sensitive data to make it anonymous, and assigns another identifying information that serves as a pseudonym. "Deletion" is a processing rule that deletes a specific attribute of the sensitive data. "Generalization" is a processing rule that changes a specific attribute of the sensitive data to a more general (higher conceptual) name. "No processing allowed" is a processing rule that does not allow processing of a specific attribute of the sensitive data.

[0021] The signature value table 309 stores pairs of identification information of the sensitive data table 305 and signature values ​​for all sensitive data in the sensitive data table 305. The signature key 310 is key information for encrypting a hash value for the sensitive data table 305. The signature key 310 is, for example, a private key of a public key cryptosystem.

[0022] <Sensitive Data Table 305> 4 is an explanatory diagram showing an example of the sensitive data table 305. The sensitive data table 305 has, as attribute fields, an ID 401, a name 402, an address 403, an age 404, and a gender 405. A combination of attribute values ​​in each field in the same row becomes an entry that specifies the sensitive data of an individual.

[0023] ID 401 is identification information that uniquely identifies the sensitive data of an individual. Name 402 is the first and last name of the individual in the sensitive data identified by ID 401. Address 403 is the place where the individual in the sensitive data identified by ID 401 resides. Age 404 is the number of years since the birth of the individual in the sensitive data identified by ID 401. Age 404 is updated over time. Gender 405 is information that distinguishes the sex of the individual in the sensitive data identified by ID 401.

[0024] <Random Number Data Table 306> 5 is an explanatory diagram showing an example of the random number data table 306. The random number data table 306 is a group of tables (541-544) that define random numbers 501-504 for each of the attributes defined in the sensitive data table 305, that is, ID 401, name 402, address 403, and age 404. The combination of the attribute and the random number value in each of the tables 541-544 is called random number data. The random numbers 501-504 are values ​​generated for each attribute (ID 401, name 402, address 403, age 404, gender 405) for which the attribute processing rule 308 is set to any one of "pseudonymization", "deletion", and "generalization".

[0025] <Generalization rule table 307> FIG. 6 is an explanatory diagram showing an example of the generalization rule table 307. The generalization rule table 307 has fields 601 before generalization and 602 after generalization. The 601 before generalization is a place name before generalization, and the 602 after generalization is a place name that includes the place name defined in the 601 before generalization and is broader than the place name. For example, if the place name in the 601 before generalization is a prefecture name, the 602 after generalization is a local area name. Also, if the place name in the 601 before generalization is a city, ward, town, or village name, the 602 after generalization is a prefecture name. For example, conversion from Tokyo to the Kanto region is a one-stage generalization. Also, conversion from Tokyo to Japan is a two-stage generalization because conversion from Tokyo to the Kanto region and conversion from the Kanto region to Japan are performed.

[0026] 6, the generalization rules are defined using place names as an example, but other information may be used instead of place names. For example, in the case of age, if the age before generalization 601 is 0-4 years old, the after generalization 602 is "child", if the age before generalization 601 is 5-14 years old, the after generalization 602 is "boy", if the age before generalization 601 is 15-24 years old, the after generalization 602 is "youth", if the age before generalization 601 is 25-44 years old, the after generalization 602 is "prime age", if the age before generalization 601 is 45-64 years old, the after generalization 602 is "middle-aged", and if the age before generalization 601 is 65 years old or older, the after generalization 602 is "elderly".

[0027] <Example of Functional Configuration of the Anonymized Data Providing Server 102> 7 is a block diagram showing an example of a functional configuration of the anonymized data providing server 102. The anonymized data providing server 102 has a Web server function 701, an anonymization processing unit 702, a hash value generating unit 303, a hash value aggregating unit 304, a sensitive data table 305, a random number data table 306, a generalization rule table 307, an attribute processing rule 308, a signature value table 309, an anonymized data table 705, a verification random number data table 706, and a verification generalization rule table 707.

[0028] The Web server function 701 and the anonymization processing unit 702 are specifically realized, for example, by causing the processor 201 to execute a program stored in the storage device 202 shown in Fig. 2. Also, the anonymized data table 705, the verification random number data table 706, and the verification generalization rule table 707 are stored in the storage device 202.

[0029] The Web server function 701 holds a Web page including identification information of the sensitive data table 305 and signature values ​​for all sensitive data in the sensitive data table 305 so as to be accessible from the anonymized data user terminal 103. The anonymization processing unit 702 executes verifiable anonymization processing on the sensitive data.

[0030] The anonymized data table 705 stores anonymized data that is the result of anonymization processing of sensitive data by the anonymization processing unit 702. The verification random number data table 706 stores random number data required for signature verification among the random number data. The verification random number data table 706 is a set of records that are obtained by duplicating attributes (ID 401, name 402, address 403, age 404, gender 405) that were not anonymized by the anonymization processing among the records of the random number data table 306 and the random value values ​​of the attribute values. The verification generalization rule table 707 stores generalization rules required for signature verification among the generalization rule table 307.

[0031] <Anonymized Data Table 705> 8 is an explanatory diagram showing an example of an anonymized data table 705. The anonymized data table 705 has fields of ID 801, name 802, address 803a, address_random number 803b, age 804, and gender 805. A combination of values ​​of each field in the same row becomes an entry that specifies the anonymized data of one individual.

[0032] ID 801 is identification information that uniquely identifies an individual's anonymized data. Name 802 is a hash value indicating the pseudonym of the individual's name 402 in the anonymized data identified by ID 801. Address 803a is the place name 602 after generalization of address 403. Address_random number 803b is a random value required for signature verification that is added to address 803a (place name after generalization 602). Age 804 is a hash value of the individual's age 404 in the anonymized data identified by ID 801 or an age to be deleted by anonymization processing (for example, under 20 years old). Gender 805 is information that distinguishes the sex of the individual in the anonymized data identified by ID 801.

[0033] <Example of Functional Configuration of the Anonymized Data User Terminal 103> 9 is a block diagram showing an example of a functional configuration of the anonymized data user terminal 103. The anonymized data user terminal 103 has a web browser function 901, a signature verification unit 902, a hash value generation unit 303, a hash value aggregation unit 304, an anonymized data table 705, a verification random number data table 706, a verification generalized rule table 707, a signature value 908, and a verification key 909.

[0034] Specifically, the Web browser function 901 and the signature verification unit 902 are realized by, for example, causing the processor 201 to execute a program stored in the storage device 202 shown in Fig. 2. Also, a signature value 908 and a verification key 909 are stored in the storage device 202.

[0035] A Web browser function 901 receives a Web page published by the anonymized data providing server 102 and displays it on the anonymized data user terminal 103. A signature verification unit 902 verifies the validity of the anonymization using a signature verification key, signature values ​​for all sensitive data in the sensitive data table 305, and the anonymized data table 705 as input.

[0036] The anonymized data table 705 stores the anonymized data acquired from the anonymized data providing server 102. The verification random number data table 706 stores the verification random number data acquired from the anonymized data providing server 102. The verification generalization rule table 707 stores the verification generalization rules acquired from the anonymized data providing server 102.

[0037] The signature value 908 is a signature value for all sensitive data in the sensitive data table 305 acquired from the anonymized data providing server 102. The verification key 909 is key information used to decrypt the signature value 908, and is, for example, a public key of a public key system for the signature key 310.

[0038] Some or all of the programs and data stored in the signature generator terminal 101, the anonymized data providing server 102, and the anonymized data user terminal 103 may be stored in advance in a non-temporary area of ​​the storage device 202 provided in the computer 200, or may be stored in the non-temporary area of ​​the storage device 202 from a non-temporary storage device of another device connected to the network 104, or from a non-temporary storage medium connected to an interface not shown, as necessary.

[0039] <Sequence of anonymization system 100> 10 is a sequence diagram of the anonymization system 100. The signature generator terminal 101 sets processing rules for each attribute of sensitive data (ID 401, name 402, address 403, age 404, gender 405) through the operation of the signature generator (step S1001). Specifically, for example, the signature generator terminal 101 displays a processing rule setting screen for each attribute of sensitive data on its output device 204, and accepts a selection of any one of "no processing", "pseudonymization", "delete" and "generalization" for each attribute using the input device 203.

[0040] 11 is an explanatory diagram showing an example of a processing rule setting screen. The processing rule setting screen 1100 has a processing rule setting area 1101, a generalization rule input area 1102, and an execute button 1103. The processing rule setting area 1101 has radio buttons for each attribute to select "Do not process", "Kanonymize", "Delete", or "Generalize", and any one of the processing rules can be selected for each attribute.

[0041] The generalization rule input area 1102 has an input field for inputting the storage destination of the generalization rule for each attribute, and a selection button for calling up a file management tool for inputting the storage destination into the input field. For example, assume that the processing rule for ID 401 is set to "pseudonymization", the processing rules for name 402 and age 404 are set to "delete", the processing rule for address 403 is set to "generalize", and the processing rule for gender 405 is set to "unprocessable".

[0042] Pressing the execute button 1103 triggers associating the processing rules set in the processing rule setting area 1101 and the generalization rule input area 1102 with attributes and storing them in the storage device 202 as attribute processing rules 308. Pressing the execute button 1103 starts random number data generation (step S1002) and signature generation processing (step S1003).

[0043] Returning to FIG. 10, the signature generator terminal 101 generates random number data for each attribute using the random number data generation unit 301, and stores the data in the random number data table 306 (step S1002). Specifically, for example, the random number data generation unit 301 performs the following process for all attributes of the sensitive data. First, the random number data generation unit 301 acquires the processing rule for the attribute from the attribute processing rule 308. If the processing rule for the attribute is "no processing allowed", the random number data generation unit 301 does not generate random number data. If the processing rule for the attribute is any of "pseudonymization", "deletion", and "generalization", the random number data generation unit 301 generates random number data.

[0044] The random number data generation unit 301 acquires attribute values ​​for each attribute from the sensitive data. For example, in the case of the sensitive data table 305 of FIG. 4, the random number data generation unit 301 acquires the attribute values ​​"247", "500", "294" and "80" of the column of ID401 for ID401. Next, the random number data generation unit 301 generates random number values ​​"8AAE34", "A7132F", "D89204" and "FB33D9" for each of the acquired attribute values ​​"247", "500", "294" and "80" of ID401, and stores the pairs of the attribute value and the random number value as random number data in the random number data table 306. The same applies to the other attributes (name 402, address 403, age 404). In this example, the processing rule for gender 405 is set to "unprocessable", so no random number value is generated for gender 405.

[0045] Next, the signature generator terminal 101 performs a signature generation process (step S1003) in which the signature generation unit 302 receives the sensitive data, the random number data, the generalization rule, the processing rule, and the signature key 310 and generates the signature value 908 for all the sensitive data in the sensitive data table 305. Details of the signature generation process (step S1003) will be described later with reference to Figs. 13 to 16.

[0046] Next, the signature generator terminal 101 transmits the sensitive data, the random number data, the generalization rule, the processing rule, and the signature value 908 generated by the signature generation process (step S1003) to the anonymized data providing server 102 (step S1004). The signature generator terminal 101 stores pairs of identification information of the sensitive data table 305 and signature values ​​for all sensitive data in the sensitive data table 305 in the signature value table 309.

[0047] The anonymized data providing server 102 acquires the sensitive data, random number data, generalization rule, processing rule, and signature value 908 transmitted from the signature generator terminal 101, and stores them in the storage device 202 (step S1005).

[0048] Next, the anonymized data providing server 102 generates a Web page accessible via the network 104 by the Web server function 701, and notifies the URL of the Web page to the Web browser function 901 of the anonymized data user terminal 103 (step S1006).

[0049] Next, the anonymized data user terminal 103 uses the web browser function 901 to obtain a web page including identification information of the sensitive data table 305, attributes of the sensitive data table 305, processing rules for each attribute of the sensitive data table 305, and signature values ​​for all sensitive data in the sensitive data table 305 (step S1007).

[0050] Next, the anonymized data user terminal 103 downloads and acquires the signature values ​​908 for all sensitive data in the sensitive data table 305 from the Web page by operating the data user's input device 203, and stores them in the storage device 202 of the anonymized data user terminal 103 (step S1008).

[0051] Next, the anonymized data user terminal 103 sets an anonymization request by operating the data user's input device 203, and transmits an anonymized data acquisition request including the anonymization request to the Web server function 701 of the anonymized data providing server 102 (step S1009). In this embodiment, the anonymized data acquisition request includes the following four anonymization requests.

[0052] Anonymization request 1: Pseudonymize ID401 Anonymity request 2: Delete name 402 Anonymization request 3: Generalize address 403 by one level Anonymity request 4: Delete if age 404 is under 19

[0053] 12 is an explanatory diagram showing an example of an anonymization request setting screen. The anonymization request setting screen 1200 is displayed on the output device 204 of the anonymized data user terminal 103, and has a processing / non-processing designation area 1201 for designating whether or not to process an attribute, a detailed request designation area 1202 for designating a detailed anonymization request for the attribute, and an execute button 1203.

[0054] The processing / non-processing designation area 1201 has radio buttons that allow the selection of the processing rule for each attribute set in the processing rule setting (step S1001). However, if the processing rule is "generalize", a "delete" radio button is also provided in addition to "generalize". Also, if the processing rule is "no processing", no radio button is provided.

[0055] The detailed request specification area 1202 is an area where detailed processing conditions can be specified when processing attributes. For example, "1 step" is specified for generalization of the address 403, requesting that the address 403 be generalized by one step. Also, "-19" is specified for deletion of the age 404, requesting that age 404 be deleted if it is 19 or younger.

[0056] When the execute button 1203 is pressed, an anonymized data acquisition request including an anonymization request for each input attribute is transmitted to the anonymized data providing server 102 (step S1009).

[0057] 10, the anonymized data providing server 102 acquires the identification information of the sensitive data table 305 and the anonymization request notified from the anonymized data user terminal 103 by the Web server function 701, and passes them to the anonymization processing unit 702. Then, the anonymized data providing server 102 executes anonymization processing by using the identification information of the sensitive data table 305 and the anonymization request as input by the anonymization processing unit 702, generates anonymized data, verification random number data, and verification generalization rules, and passes them to the Web server function 701 (step S1010). Details of the anonymization processing (step S1010) will be described later with reference to FIGS. 17 to 19.

[0058] The Web server function 701 registers the anonymized data, the verification random number data, and the verification generalization rule on a download Web page for data users, and notifies the Web browser function 901 of the anonymized data user terminal 103 of the URL.

[0059] Next, the web browser function 901 of the anonymized data user terminal 103 uses the notified URL as input to access a download web page for data users, downloads and acquires the anonymized data, the verification random number data, and the verification generalization rule, and stores them in the storage device 202 of the anonymized data user terminal 103 (step S1011). As a result, an anonymized data table 705, a verification random number data table 706, and a verification generalization rule table 707 are created in the storage device 202 of the anonymized data user terminal 103.

[0060] Finally, the anonymized data user terminal 103 executes a signature verification process (step S1012) to verify the legitimacy of the anonymization process (step S1010) performed by the anonymized data providing server 102, using the anonymized data, the verification random number data table 706, the verification generalized rule table 707, the signature value 908, and the verification key 909 as inputs, by the signature verification processing unit 317. If the legitimacy of the anonymization process (step S1010) is verified, the anonymized data user terminal 103 displays "verification successful" on the output device 204, and if the legitimacy of the anonymization process (step S1010) is not verified, displays "verification failed" on the output device 204 (step S1012).

[0061] If the verification is successful, the anonymized data user terminal 103 displays the anonymized data on the output device 204, and if the verification is unsuccessful, the anonymized data is not displayed on the output device 204. Details of the signature verification process (step S1012) will be described later with reference to FIG. 20.

[0062] 10, an anonymized data user can verify whether the anonymized data he or she obtained has been legitimately anonymized through legitimate anonymization process (step S1010) that follows the anonymization rules set by the signature generator. This makes it possible to prevent the provision of services that use fraudulent analysis results obtained from fraudulent anonymized data.

[0063] <Signature Generation Process (Step S1003)> Fig. 13 is a flowchart showing a detailed example of the processing procedure of the signature generation process (step S1003) shown in Fig. 10. The signature generation unit 302 generates a hash value (hereinafter, cell hash value) HC for each cell of sensitive data (step S1301). The sensitive data table 305 is a table with N records (N=4 in Fig. 4) and M attributes (M=5 in Fig. 4), and therefore has MxN cells (20 in Fig. 4). Therefore, the signature generation unit 302 generates 20 cell hash values ​​HC. Details of this cell hash value generation process (step S1301) will be described later with reference to Fig. 14.

[0064] Next, the hash value aggregating unit 304 generates a hash value (record hash value) HR for each record in the sensitive data table 305 from the cell hash value HC (step S1302). Specifically, the signature generating unit 302 sequentially reads out the N records (N=4 in FIG. 4) of the cell hash value HC, and passes them to the hash value aggregating unit 304. The hash value aggregating unit 304 generates a record hash value HR by concatenating the cell hash values ​​HC of M attributes for each record. That is, N record hash values ​​HR are generated. Details of this record hash value generation process (step S1302) will be described later with reference to FIG. 16.

[0065] Next, the hash value aggregation unit 304 combines the record hash values ​​HR for each record to generate a hash value (hereinafter, total hash value) HT for the entire sensitive data table 305 (step S1303). Details of this total hash value generation process (step S1303) will be described later with reference to FIG.

[0066] Finally, the signature generating unit 302 generates a signature value 908 using the total hash value HT and the signature key 310 (step S1304).

[0067] <Cell hash value generation process (step S1301)> FIG. 14 is a flowchart illustrating a detailed example of the cell hash value generation process (step S1301) illustrated in FIG.

[0068] The signature generation unit 302 initializes a variable i (hereinafter, may be written as attribute i) indicating the number of an attribute in the sensitive data table 305 to 1 (for example, i=1 is ID401) (step S1401). Next, the signature generation unit 302 determines whether i>M is satisfied (step S1402). If i>M is satisfied (step S1402: Yes), the signature generation unit 302 ends the cell hash value generation process (step S1101) and proceeds to step S1302.

[0069] On the other hand, if i>M is not true (step S1402: No), the signature generating unit 302 determines whether the processing rule for the i-th attribute corresponds to either "delete" or "generalize" (step S1403).

[0070] If the processing rule for the i-th attribute does not correspond to either "delete" or "generalize" (step S1403: No), the process proceeds to step S1405. For example, in this example, the ID 401, the name 402, and the gender 405 do not correspond to either the processing rule "delete" or the processing rule "generalize".

[0071] On the other hand, address 403 and age 404 correspond to either "delete" or "generalize" (step S1403: Yes). In this case, signature generating unit 302 generates a counter (step S1404).

[0072] Specifically, for example, the signature generation unit 302 first obtains an attribute value for the i-th attribute (address 403 or age 404) from the random number data table 306. For example, if the attribute i is the address 403, the signature generation unit 302 obtains "Tokyo", "Kanagawa", and "Osaka" as the attribute values ​​xi, and if the attribute i is the age 404, the signature generation unit 302 obtains "16", "37", "48", and "50" as the attribute values ​​xi. Then, the signature generation unit 302 generates a counter for each attribute value xi, and initializes the counter value ci of each counter to 0. After S1404, the process proceeds to S1405.

[0073] After generating the counter (step S1404), the signature generating unit 302 initializes a variable j (hereinafter, may be written as record j) indicating the ordinal number of the record in the sensitive data table 305 to 1 (step S1405).

[0074] Then, the signature generating unit 302 judges whether j>N is satisfied (step S1406). If j>N is not satisfied (step S1406: No), the signature generating unit 302 generates a cell hash value HCij of the attribute value xij in the cell Cij of the attribute i and record j (step S1407). For example, if the attribute i is i=1 and the record j is j=1, the attribute value xij is "247" of the ID 401 in the record in the first row. The cell hash value HCij generation process (step S1407) generates the cell hash value HCij of the attribute value xij according to the processing rule for the attribute i.

[0075] Specifically, for example, the process of generating the cell hash value HCij when the processing rule for the attribute i is "generalized" (step S1407) will be described later with reference to FIG.

[0076] The process of generating the cell hash value HCij when the processing rule for the attribute i is “delete” (step S1407) is the same as part of the process when the processing rule for the attribute i is “generalize”, so will be explained together in Figure 15 (steps S1501 to S1505).

[0077] When the processing rule for the attribute i is "pseudonymization," the signature generation unit 302 first obtains a random value rij for the attribute value xij from the random number data, inputs a value xijrij obtained by concatenating the attribute value xij and the random value rij to the hash value generation unit 303, and sets the output result from the hash value generation unit 303 as the cell hash value HCij of the cell Cij. As a result, multiple identical attribute values ​​xij become the same concatenated value xijrij and also the same cell hash value HCij.

[0078] If the processing rule for the attribute i is "unprocessable", the signature generation unit 302 inputs the attribute value xij to the hash value generation unit 303, and sets the output result from the hash value generation unit 303 as the cell hash value HCij of the cell.

[0079] After step S1407, signature generating unit 302 increments variable j (step S1408) and returns to step S1406. If j>N is satisfied in step S1406 (step S1406: Yes), the process proceeds to step S1409 because the process has been completed up to the final record j=N for attribute i.

[0080] Then, the signature generation unit 302 increments the variable i (step S1409) and returns to step S1402. If i>M in step S1402 (step S1402: Yes), the signature generation unit 302 ends the cell hash value generation process (step S1301) because the process has been completed up to the last attribute i=M, and proceeds to step S1302. As a result, cell hash values ​​HCij for all cells Cij in the sensitive data table 305 are generated.

[0081] <Generation process of cell hash value HCij (step S1407)> Fig. 15 is a flowchart showing a detailed example of the process of generating the cell hash value HCij (step S1407) shown in Fig. 14. Fig. 15 shows a detailed example of the process of generating the cell hash value HCij (step S1407) when the processing rule for the attribute i is "generalize" or "delete".

[0082] The signature generation unit 302 acquires the counter value ci generated in step S1404 for the target record j (step S1501). Next, the signature generation unit 302 acquires the random number rij for the attribute value xij from the random number data table 306 (step S1502). Next, the signature generation unit 302 calculates the sum r'ij=rij+ci of the counter value ci and the random value rij (step S1503).

[0083] Next, the signature generation unit 302 increments the counter value ci (step S1504), inputs the value xijr'ij obtained by concatenating the attribute value xij and the random value r'ij to the hash value generation unit 303, and obtains the hash value H that is the output result from the hash value generation unit 303 (step S1505).

[0084] Then, the signature generating unit 302 judges whether the attribute value xij is generalizable or not (step S1506). Specifically, for example, the signature generating unit 302 determines that the attribute value xij is generalizable if it exists in the pre-generalization 601 of the generalization rule table 307. If the attribute value xij is generalizable (step S1506: Yes), the signature generating unit 302 updates the attribute value xij and the random value r'ij (step S1507) and returns to step S1505.

[0085] Specifically, for example, the signature generation unit 302 obtains the generalized value 602 of the attribute value xij from the generalization rule table 307, and updates the attribute value xij. Also, the signature generation unit 302 sets the hash value H obtained in step S1505 to the random value r'ij of the updated attribute value xij.

[0086] In step S1506, if the hash value is not generalizable (step S1506: No), the signature generation unit 302 sets the last generated hash value H to the cell hash value HCij of the attribute value xij (step S1508), and the process of generating the cell hash value HCij (step S1407) ends.

[0087] In this way, a cell hash value HCij for the attribute value xij of the attribute i whose processing rule is "generalize" can be generated according to the number of stages of generalization.

[0088] Also, when the processing rule is "delete", the signature generation unit 302 executes steps S1501 to S1505, and sets the cell hash value H generated in step S1505 as the cell hash value HCij of the attribute value xij. As a result, even if there are multiple identical attribute values ​​xij, they will have different linked values ​​xijrij and different cell hash values ​​HCij, preventing the deleted attribute value xij from being guessed.

[0089] <Hash value aggregation process> 16 is a flowchart showing a detailed example of a procedure for the hash value aggregation process. The hash value aggregation process is a process for aggregating multiple hash values ​​to generate one hash value, and is applied to the record hash value generation process (step S1302) and the total hash value generation process (step S1303).

[0090] The multiple hash values ​​before aggregation are multiple cell hash values ​​HCij when applied to the record hash value generation process (step S1302), and are multiple record hash values ​​HRj when applied to the whole hash value generation process (step S1303). In Fig. 16, each of the multiple hash values ​​before aggregation is called an "element".

[0091] The hash value aggregator 304 initializes variables k and l to 1, sets a list containing multiple elements as element list E, initializes variable m to the number of elements in element list E (set as E), and generates an empty list E' (step S1601).

[0092] Next, the hash value aggregation unit 304 determines whether the variable m is equal to 1 (step S1601). If the variable m is not equal to 1 (step S1602: No), the hash value aggregation unit 304 generates a hash value by concatenating the k-th element Ek and the (k + 1)-th element Ek+1 in the element list E, and sets it as the l-th element E’l of the list E’ (step S1603).

[0093] Next, the hash value aggregation unit 304 increments the value of k by 2 and the value of l by 1 (step S1603). Next, the hash value aggregation unit 304 determines whether k < m (step S1605). If k < m (step S1605: Yes), it returns to step S1602. On the other hand, if k is not less than m (step S1605: No), the hash value aggregation unit 304 determines whether k = m (step S1606). If k is not equal to m (step S16063: No), it proceeds to step S1608, and if k = m (step S1067: Yes), the hash value aggregation unit 304 adds the value of the element Ek to the list E’ as the l-th element of E’, and proceeds to step S1608.

[0094] The hash value aggregation unit 304 sets the variable k to k = 1, the variable l to l = 1, the variable m to the number of elements in the list E’, sets the number of elements E in the element list E to E = E’, and sets E’ to an empty list, and returns to S1602. In step S1602, if the variable m is equal to 1 (step S1602: Yes), the hash value aggregation unit 304 sets the aggregated hash value H to E1. Through the above processing, one hash value E1 obtained by aggregating a plurality of hash values is generated.

[0095] That is, when applied to the record hash value generation process (step S1302), the hash value E1 is the record hash value HRj of a certain record j, and when applied to the overall hash value generation process (step S1303), the hash value E1 is the overall hash value.

[0096] In this way, the signature generator terminal 101 can generate the total hash value HT by making the attribute values ​​xij of all the cells in the sensitive data table 305 into cell hash values.

[0097] <Anonymization process (step S1010)> Fig. 17 is a flowchart showing a detailed example of the processing procedure of the anonymization process (step S1010) shown in Fig. 10. The anonymization process (step S1010) is a process in which the anonymization processing unit 702 of the anonymization data providing server 102 anonymizes sensitive data by using sensitive data, random number data, generalization rule data, and processing rules, so that the validity of the anonymization process (step S1010) can be verified.

[0098] The anonymization processor 702 initializes a variable i to 1 (step S1701). Next, the anonymization processor 702 judges whether i>M is satisfied (step S1702). If i>M is not satisfied (step S1702: No), the anonymization processor 702 judges whether processing of the attribute i is requested in step S1009 (step S1703). If processing of the attribute i is not requested (step S1703: No), the anonymization processor 702 copies the attribute value xi of the attribute i to the attribute i in the anonymized data table 705 (step S1706) and proceeds to step S1707.

[0099] On the other hand, if processing of the attribute i is requested (step S1703: Yes), anonymization processing is performed on the attribute value xi of the attribute i (step S1704). Details of the anonymization processing (step S1704) will be described later with reference to FIG.

[0100] Then, the anonymization processing unit 702 generates verification random number data (step S1705). Specifically, for example, the anonymization processing unit 702 first refers to the anonymization processing request for the attribute i, and determines whether or not the anonymization processing request is a request for all attribute values ​​xi of the attribute i. If the request is for processing all attribute values ​​xi, the anonymization processing unit 702 does not generate verification random number data. For example, if the attribute i is the ID 401, the name 402, or the address 403, the request is for processing all attribute values ​​xi, and therefore the anonymization processing unit 702 does not generate verification random number data.

[0101] On the other hand, if the processing request is for a part of the attribute values ​​xi, the anonymization processor 702 generates random number data for verification. For example, if the attribute i is the age 404, the processing request is to delete those aged 19 or younger.

[0102] Specifically, the anonymization processor 702 first obtains random number data of the attribute i from the random number data table 306 and duplicates it as verification random number data. For example, the anonymization processor 702 obtains random number data that is a combination of the age 404 and the random number 504 from the table 544 of the random number data table 306 and duplicates it as verification random number data.

[0103] Next, the anonymization processor 702 refers to the anonymization request (deletion) for the attribute i, determines the random number data to be deleted from the verification random number data, and finally deletes the random number data to be deleted from the verification random number data to obtain the final verification random number data. For example, the anonymization processor 702 determines the value of age 404 that is 19 years old or younger and the value of random number 504 in the verification random number data to be deleted, and deletes them from the verification random number data group to obtain the final verification random number data group. That is, the values ​​of age 404 "37", "48" and "50" and the values ​​of random number 504 "BA5CBF", "76E635" and "1C104A" become the final verification random number data. Then, the process proceeds to step S1707.

[0104] The anonymization processor 702 increments the variable i (step S1707), and returns to step S1702. If i>M is satisfied in step S1702 (step S1702: Yes), the anonymization process (step S1010) ends.

[0105] <Anonymization processing (step S1704)> Fig. 18 is a flowchart showing a detailed example of a processing procedure of the anonymization processing (step S1704) shown in Fig. 17. The anonymization processor 702 determines whether the processing rule for the attribute i is "generalization" (step S1801).

[0106] If the processing rule for attribute i is “generalize” (step S1801: Yes), the anonymization processor 702 adds the random number sequence for attribute i to the anonymized data table 705 (step S1802), and proceeds to step S1804. For example, if attribute i is address 403, the anonymization processor 702 adds the random value sequence for random number 503 to the anonymized data table 705.

[0107] On the other hand, if the processing rule for attribute i is not "generalize" (step S1801: No), the anonymization processor 702 judges whether the processing rule for attribute i is "delete" (step S1803). If the processing rule for attribute i is "delete" (step S1803: Yes), the process proceeds to step S1804. On the other hand, if the processing rule for attribute i is not "delete" (step S1803: No), the process proceeds to step S1805.

[0108] After step S1802 or step S1803: Yes, the anonymization processor 702 generates a counter (step S1804) and proceeds to step S1805. That is, when the processing rule for attribute i is "generalize" or "delete", a counter is generated for each attribute value xi of attribute i. The counter value c of each counter is initialized to 0.

[0109] The anonymization processor 702 sets a variable j, which indicates which record in the sensitive data table 305, to 1 (step S1805), and judges whether j>N is satisfied (step S1806). If j>N is not satisfied (step S1806: No), the anonymization processor 702 judges whether processing is required for the attribute value xij (step S1807).

[0110] If processing is requested for the attribute value xij (step S1807: Yes), the anonymization processor 702 executes anonymization processing of the attribute value xij in response to the processing request (step S1808), and proceeds to step S1810. The anonymization processing (step S1808) when the processing request is "generalization" will be described later with reference to FIG.

[0111] The anonymization process (step S1808) when the processing request is “delete” is similar to S1501 to S1505 in FIG.

[0112] The anonymization process (step S1808) when the processing request is “pseudonymization” is the same as the generation process of the cell hash value HCij when the processing rule is “pseudonymization” (step S1407), and the generated cell hash value HCij is stored in the anonymized data table 705.

[0113] On the other hand, if processing of the attribute value xij is not requested (step S1807: No), the anonymization processor 702 copies the attribute value xij of the attribute i to the attribute i of the anonymized data table 705 (step S1809), and proceeds to step S1810.

[0114] The anonymization processor 702 increments the variable j (step S1810) and returns to step S1806. If j>N is satisfied in step S1806 (step S1806: Yes), the anonymization processor 702 ends the anonymization processing (step S1704) and proceeds to step S1705.

[0115] <Anonymization process of attribute value xij (step S1808)> Fig. 19 is a flowchart showing a detailed example of the procedure of the anonymization process (step S1808) of the attribute value xij when the processing request shown in Fig. 18 is "generalize". In Fig. 19, the same processes as those in Fig. 15 are denoted by the same step numbers.

[0116] After step S1503, the anonymization processor 702 initializes a variable l to 0 (step S1904). Next, the anonymization processor 702 determines whether the variable l matches the generalization level of the processing request (step S1905). The generalization level is the number of conversions from pre-generalization 601 to post-generalization 602, such as one stage, two stages, .... For example, when generalizing from Chiyoda-ku to Tokyo, Tokyo to Kanto region, and Kanto region to Japan, the generalization level is three stages. In this example, steps S1505, S1507, and S1908 are repeated three times.

[0117] If the variable l matches the generalization level of the processing request (step S1905: Yes), the anonymization processor 702 executes steps S1105 and S1107, increments the variable l (step S1908), and returns to step S1905.

[0118] If the variable l does not match the generalization level of the processing request (step S1905: No), the anonymization processing unit 702 stores the attribute value xij in cell Ciaj of record j of attribute ia (e.g., Address 803a) in the anonymized data table 705, and stores the hash value H in cell Cibj of record j of the random number sequence ib (e.g., Address_Random Number 803b) for attribute i.

[0119] Then, the anonymization processor 702 generates a verification generalization rule table 707 (step S1910). Specifically, for example, the anonymization processor 702 first copies the generalization rule of the attribute i to an empty verification generalization rule table 707. Next, the anonymization processor 702 refers to the processing request of the attribute i and the generalization rule table 307, and determines a record to be deleted from the generalization rule table 307.

[0120] Specifically, for example, in the case of a processing request to generalize the address 403 by three levels, the anonymization processor 702 identifies a record that generalizes the attribute value xij indicating the address 403 by three levels from the generalization rule table 307. For example, if the attribute value xij indicating the address 403 is "Chiyoda-ku", a record before generalization 601 (Chiyoda-ku) and after generalization 602 (Tokyo), a record before generalization 601 (Tokyo) and after generalization 602 (Kanto region), and a record before generalization 601 (Kanto region) and after generalization 602 (Japan) are identified.

[0121] The anonymization processor 702 determines, among the identified records, the records before and after generalization other than the final stage as records to be deleted. In this example, the records before generalization 601 (Chiyoda-ku) and after generalization 602 (Tokyo) and the records before generalization 601 (Tokyo) and after generalization 602 (Kanto region) are determined as records to be deleted. The anonymization processor 702 generates the generalization rule table 307 after deletion of the records to be deleted as a verification generalization rule table 707.

[0122] In this way, the anonymized data providing server 102 generates cell hash values ​​only for the attribute values ​​xij of the cell cij for which anonymization has been requested in the sensitive data table 305, and replaces the cell hash values ​​generated from the original attribute values ​​xij, thereby generating the anonymized data table 705. The remaining attribute values ​​that have not been anonymized are converted to cell hash values ​​in the anonymized data user terminal 103.

[0123] <Signature Verification Process (Step S1012)> Fig. 20 is a flowchart showing a detailed example of the processing procedure of the signature verification process (step S1012) shown in Fig. 10. The anonymized data user terminal 103 executes an anonymized total hash value generation process for all anonymized data in the anonymized data table 705 acquired in step S1011 (step S2001). In the anonymized total hash value generation process (step S2001), a hash value H' of the entire anonymized data table 705 (hereinafter, anonymized total hash value) is generated. Details of the anonymized total hash value generation process (step S2001) will be described later with reference to Fig. 21.

[0124] The anonymized data user terminal 103 decrypts the signature value 908 for all sensitive data in the sensitive data table 305 with the verification key 909 to obtain the total hash value HT (step S2002). Then, the anonymized data user terminal 103 determines whether the decrypted total hash value HT matches the anonymized total hash value H' (step S2003).

[0125] If they match (step S2003: Yes), the anonymized data user terminal 103 displays "Verification successful" on the output device 204 (step S2004). On the other hand, if they do not match (step S2003: No), the anonymized data user terminal 103 displays "Verification failed" on the output device 204 (step S2005), and the signature verification process (step S1012) ends.

[0126] <Anonymized overall hash value generation process> Fig. 21 is a flowchart showing a detailed example of the process steps of the anonymized global hash value generation process (step S2001) shown in Fig. 20. The anonymized data user terminal 103 initializes a variable i to 1 (step S2101).

[0127] The anonymized data user terminal 103 determines whether i>M is satisfied (step S2102). If i>M is not satisfied (step S2102: No), the anonymized data user terminal 103 determines whether the attribute i is a "random number" (step S2103). For example, if the attribute i is address_random number 803b, the attribute i is determined to be a "random number". If the attribute i is a "random number" (step S2103: Yes), the process proceeds to step S2114.

[0128] On the other hand, if the attribute i is not "random number" (step S2103: No), the anonymized data user terminal 103 judges whether or not verification random number data for the attribute i exists in the verification random number data table 706 (step S2104). If there is no verification random number data for the attribute i and there is no random number sequence for the attribute i in the anonymized data table 705 (step S2104: No), the process proceeds to step S2105. If there is verification random number data for the attribute i or if there is a random number sequence for the attribute i in the anonymized data (step S2104: Yes), the process proceeds to step S2106.

[0129] In step S2105, the anonymized data user terminal 103 sets the cell hash value of cell cij of attribute i of the anonymized data as the cell hash value of the anonymized data (step S2105), and proceeds to step S2114. For example, for the cell hash value "4B4568" (see FIG. 8) obtained by anonymizing the attribute value xij of age 404 due to a processing request (deletion), the attribute value does not exist in the verification random number data table 706, so the cell hash value "4B4568" is used as is.

[0130] In step S2106, the anonymized data user terminal 103 determines whether the processing rule for attribute i is "delete" or "generalize" (step S2106). If the processing rule for attribute i is not "delete" or "generalize" (step S2106: No), the process proceeds to step S2108. On the other hand, if the processing rule for attribute i is "delete" or "generalize" (step S2106: Yes), the anonymized data user terminal 103 generates a counter (step S2107).

[0131] Specifically, for example, the anonymized data user terminal 103 first acquires an attribute value for the i-th attribute (address 403 or age 404) from the verification random number data table 706. For example, if the attribute i is the address 403, the anonymized data user terminal 103 acquires "Tokyo", "Kanagawa", and "Osaka" as the attribute values ​​xi, and if the attribute i is the age 404, it acquires "16", "37", "48", and "50" as the attribute values ​​xi. Then, the anonymized data user terminal 103 generates a counter for each attribute value xi, initializes the counter value ci of each counter to 0, and proceeds to step S2108.

[0132] After generating the counter (step S2107), the anonymized data user terminal 103 initializes a variable j (hereinafter, sometimes written as record j) indicating the ordinal number of the record in the anonymized data table 705 to 1 (step S2108).

[0133] Then, the anonymized data user terminal 103 judges whether j>N is satisfied (step S2109). If j>N is not satisfied (step S2109: No), the anonymized data user terminal 103 judges whether the random number data of the attribute value xij in the cell Cij of the attribute i and the record j is present in the verification random number data table 706 (step S2110).

[0134] If the random number data of the attribute value xij does not exist in the verification random number data table 706 (step S2110: No), the anonymized data user terminal 103 sets the attribute value xij to the cell hash value HCij (step S2112), and proceeds to step S2113.

[0135] On the other hand, if the random number data for the attribute value xij exists in the verification random number data table 706 (step S2110: Yes), the anonymized data user terminal 103 generates a cell hash value HCij for the attribute value xij in accordance with the attribute processing rules for attribute i (step S2111) and proceeds to step S2113.

[0136] Specifically, for example, in step S2111, the anonymized data user terminal 103 performs the same process (step S1407) for generating the cell hash value HCij shown in Fig. 15 when the processing rule for the attribute i is "generalize". However, the random number rij for the attribute value xij in step S1502 is acquired from the random number sequence (for example, address_random number 803b) in the anonymized data table 705. In addition, when applying the process (step S1407) for generating the cell hash value HCij shown in Fig. 15, the anonymized data user terminal 103 uses the verification generalization rule table 707 instead of the generalization rule table 307 in step S1506.

[0137] When the processing rule for the attribute i is "delete", the process is the same as steps S1501 to S1504 in the process of generating the cell hash value HCij shown in FIG. 15 (step S1407).

[0138] The process when the processing rule for the attribute i is "pseudonymization" or "impossible to process" is the same as the process of generating the cell hash value HCij shown in FIG. 15 (step S1407).

[0139] After that, the anonymized data user terminal 103 increments the variable j (step S2113) and returns to step S2109. If j>N is satisfied in step S2109 (step S2109: Yes), the process proceeds to step S2114. In step S2114, the anonymized data user terminal 103 increments the variable i (step S2114) and returns to step S2102.

[0140] In step S2102, if i>M (step S2102: Yes), the anonymized data user terminal 103 ends the anonymized global hash value generation process (step S2001) and proceeds to step S2002. Through the above process, an anonymized global hash value H' consisting of cell hash values ​​for all cell values ​​in the anonymized data table 705 is generated.

[0141] In this way, the anonymized data user terminal 103 sets the non-anonymized attribute value xij (not a cell hash value) in the anonymized data table 705 to a cell hash value, so that all cells in the anonymized data table 705 become cell hash values. Then, the anonymized data user terminal 103 can generate an anonymized total hash value H' from all cell hash values ​​in the anonymized data table 705. Therefore, no matter which attribute value xij is requested to be anonymized, the anonymized data user terminal 103 can legitimately generate an anonymized total hash value H' from the anonymized data table 705.

[0142] As described above, if the total hash value HT generated from the sensitive data table 305 by the signature generation process (step S1003) described in Figures 13 to 16 matches the anonymized total hash value H' generated by the signature verification process (step S1012) shown in Figure 20 for the anonymized data table 705 that has been subjected to the anonymization process (step S1010) described in Figures 17 to 19, it is possible to verify that an unauthorized anonymization process (step S1010) has not been performed, that is, the legitimacy of the anonymization.

[0143] In addition, when generating a hash value, by concatenating a random number with the original cell value and hashing it, the amount of calculations required to identify the original data from the hash value becomes enormous, thereby reducing the risk of the original data being restored from the anonymized data.

[0144] Furthermore, a random number to be concatenated to the original data is generated for each attribute value of each attribute, and if there are multiple attribute values, a new random number is generated from the random number for that attribute value and used, thereby reducing the data size of the random numbers for attributes with fewer occurrence patterns of attribute values.

[0145] To ensure security, the signature generator terminal 101 generates the signature value 908 using the total hash value HT and the signature key 310 (step S1304), but if security is guaranteed by another method, the signature generator terminal 101 may transmit the total hash value HT as it is to the anonymized data providing server 102. In this case, the anonymized data providing server 102 transmits the total hash value HT to the anonymized data user terminal 103.

[0146] In the above-described embodiment, the anonymization system is configured by the signature generator terminal 101, the anonymized data providing server 102, and the anonymized data user terminal 103, but the anonymized data providing server 102 may execute the processes (steps S1001 to S1003) executed by the signature generator terminal 101. In this case, data transmission (step S1004) and data acquisition (step S1005) become unnecessary.

[0147] In addition, the anonymization system 100 having the above-mentioned signature generator terminal 101, the anonymized data providing server 102 capable of communicating with the signature generator terminal 101, and the anonymized data user terminal 103 capable of communicating with the anonymized data providing server 102 can also be configured as follows (1) to (7).

[0148] (1) The anonymization system 100 includes a signature generation process (step S1003) in which the signature generator terminal 101 generates a second hash value (total hash value) of a plurality of data groups (sensitive data table 305) having different attributes based on a first hash value (cell hash value) of each data of the plurality of data groups; a first acquisition process (step S1005) in which the anonymized data providing server 102 acquires the plurality of data groups and the second hash value generated by the signature generation process (step S1003); an anonymization process (step S1010) in which the anonymized data providing server 102 generates a first hash value of each data of a data group having a specific attribute among the plurality of data groups in response to an anonymization request from the anonymized data user terminal 103; a second acquisition process (step S1011) in which, as a result of the anonymization request for the data group of the specific attribute being transmitted to the anonymized data providing server 102 by the anonymized data user terminal 103, the first hash value of each data of the data group of the specific attribute, data groups of remaining attributes of the plurality of data groups excluding the data group of the specific attribute, and the second hash value are acquired from the anonymized data providing server 102; an anonymized overall hash value generation process (step S2001) in which the anonymized data user terminal 103 generates a first hash value for each piece of data in the data group having the remaining attribute acquired by the second acquisition process (step S1011), and generates a third hash value based on the first hash value for each piece of data in the data group having the remaining attribute and the first hash value for each piece of data in the data group having the specific attribute acquired by the second acquisition process (step S1011); The anonymized data user terminal 103 executes a verification process (step S2003) for verifying the validity of the anonymization process (step S1010) by determining whether or not the second hash value acquired by the second acquisition process (step S1011) matches the third hash value (anonymized overall hash value) generated by the anonymized overall hash value generation process (step S2001).

[0149] (2) In the anonymization system 100 described above in (1), In the signature generation process (step S1003), the signature generator terminal 101 generates a first hash value of the data in accordance with a processing rule corresponding to the attribute, and generates second hash values ​​of the plurality of data groups based on the first hash value of the data; In the first acquisition process (step S1005), the anonymized data providing server 102 acquires the processing rule from the signature generator terminal 101, In the anonymization process (step S1010), the anonymized data providing server 102 generates a first hash value of each data of the data group having the specific attribute in accordance with the processing rule, In the second acquisition process (step S1011), the anonymized data user terminal 103 acquires the processing rule from the anonymized data providing server 102, In the anonymized entire hash value generation process (step S2001), the anonymized data user terminal 103 generates a first hash value of each data of the data group of the remaining attributes in accordance with the processing rule.

[0150] (3) In the anonymization system 100 described above in (2), the signature generator terminal 101 has a random number data table that holds random number data, which is a combination of the data and a random number value, for each of the data; In the signature generation process (step S1003), when the processing rule is to tokenize the data, the signature generator terminal 101 generates a first hash value of each data of the data group having a tokenization target attribute based on a data group having a tokenization target attribute to which the tokenization is applied and a random value of each data of the data group; In the first acquisition process (step S1005), the anonymized data providing server 102 acquires the random number data for each of the data from the signature generator terminal 101 and generates the random number data table; In the anonymization process (step S1010), if the specific attribute is the attribute to be pseudonymized, the anonymized data providing server 102 generates a first hash value for each data in the data group of the specific attribute based on each data in the data group of the specific attribute and its random value value.

[0151] (4) In the anonymization system 100 described above in (2), the signature generator terminal 101 has a random number data table that holds random number data, which is a combination of the data and a random value thereof, for each of the data; In the signature generation process (step S1003), when the processing rule is to delete the data, the signature generator terminal 101 generates a first hash value of each data of the data group having an attribute to be deleted based on each data of the data group having an attribute to be deleted to which the deletion is applied, and its random value and unique value (counter value); In the first acquisition process (step S1005), the anonymized data providing server 102 acquires the random number data for each of the data from the signature generator terminal 101 and generates the random number data table; In the anonymization process (step S1010), if the specific attribute is the attribute to be deleted, the anonymized data providing server 102 generates a first hash value for each data in the data group of the specific attribute based on each data in the data group of the specific attribute, its random number value, and the unique value.

[0152] (5) In the anonymization system 100 described above in (4), In the anonymization process (step S1010), when the anonymization request is for a portion of data of the specific attribute, the anonymized data providing server 102 generates a first hash value of the portion of data of the specific attribute, In the second acquisition process (step S1011), the anonymized data user terminal 103 transmits the anonymization request to the anonymized data providing server 102, and as a result, acquires from the anonymized data providing server 102 a first hash value of the part of data of the specific attribute, a remaining data group of the plurality of data groups excluding the part of data of the specific attribute, and the second hash value, acquires verification random number data which is a combination of other data of the specific attribute other than the part of data of the specific attribute and its random value value, and generates a verification random number data table; In the anonymized entire hash value generation process (step S2001), the anonymized data user terminal 103 generates a first hash value for each data of the remaining data group acquired by the second acquisition process (step S1011), for data groups having attributes other than the specific attribute, generates a first hash value for the data of the remaining data group acquired by the second acquisition process (step S1011), for the data group of the specific attribute, if the data is present in the verification random number data table, based on the data, its random number value, and a unique value (counter value), and generates the third hash value based on the first hash value of each data of the remaining data group and the first hash value of a part of the data of the specific attribute.

[0153] (6) In the anonymization system 100 described above in (2), the signature generator terminal 101 has a random number data table that holds random number data, which is a combination of the data and a random value thereof, for each of the data, and a generalization rule table that specifies combinations of data before and after generalization; In the signature generation process (step S1003), when the processing rule is generalization of the data, the signature generator terminal 101 generates a first hash value of each data of the data group of the attribute to be generalized based on the data of the attribute to be generalized to which the generalization is applied, its random value and unique value (counter value), and the number of conversions (generalization level) from before generalization to after generalization according to the generalization rule table, In the first acquisition process (step S1005), the anonymized data providing server 102 acquires the random number data for each of the data from the signature generator terminal 101 to generate the random number data table, acquires a combination of data before and after generalization of the generalization rule table to generate the generalization rule table, In the anonymization process (step S1010), if the specific attribute is the attribute to be generalized, the anonymized data providing server 102 generates a first hash value for each data in the data group of the specific attribute based on each data in the data group of the specific attribute, its random number value and unique value, and the number of conversions.

[0154] (7) In the anonymization system 100 described above in (1), In the anonymized whole hash value generation process (step S2001), the anonymized data providing server 102 generates a first hash value of a part of data having a specific attribute for which anonymization is requested, In the second acquisition process (step S1011), the anonymized data user terminal 103 transmits the anonymization request for the part of the data having the specific attribute to the anonymized data providing server 102, and as a result, acquires from the anonymized data providing server 102 a first hash value of the part of the data having the specific attribute, a remaining data group excluding the part of the data having the specific attribute from the multiple data groups, and the second hash value, In the anonymized entire hash value generation process (step S2001), the anonymized data user terminal 103 generates a first hash value for each data of the remaining data group acquired by the second acquisition process (step S1011), and generates the third hash value based on the first hash value of each data of the remaining attribute data group and the first hash value of a portion of the data of the specific attribute acquired by the second acquisition process (step S1011).

[0155] The present invention is not limited to the above-described embodiments, and includes various modified examples and equivalent configurations within the spirit of the appended claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those having all of the configurations described. Also, a part of the configuration of one embodiment may be replaced with a configuration of another embodiment. Also, a configuration of another embodiment may be added to a configuration of one embodiment. Also, a part of the configuration of each embodiment may be added, deleted, or replaced with another configuration.

[0156] Furthermore, each of the aforementioned configurations, functions, processing units, processing means, etc. may be realized in hardware, for example by designing some or all of them as an integrated circuit, or may be realized in software by a processor interpreting and executing a program that realizes each function.

[0157] Information such as programs, tables, files, etc. that realize each function can be stored in a storage device 202 such as a memory, a hard disk, or an SSD (Solid State Drive), or in a recording medium such as an IC (Integrated Circuit) card, an SD card, or a DVD (Digital Versatile Disc).

[0158] In addition, the control lines and information lines shown are those considered necessary for the explanation, and do not necessarily show all the control lines and information lines necessary for implementation. In reality, it can be considered that almost all components are connected to each other. [Explanation of symbols]

[0159] 100 Anonymization System 101 Signature generator terminal 102 Anonymous data provision server 103 Anonymous Data User Terminal 301 Random Number Data Generation Unit 302 Signature generation section 303 Hash value generator 304 Hash value aggregation unit 305 Sensitive Data Table 306 Random Number Data Table 307 Generalized Rules Table 308 Attribute Processing Rules 309 Signature Value Table 310 Signing Key 317 Signature Verification Processing Unit 705 Anonymized Data Tables 706 Random number data table for verification 707 Generalized rule table for validation

Claims

1. An anonymization system having a first device, a second device capable of communicating with the first device, and a third device capable of communicating with the second device, a first generation process in which the first device generates second hash values ​​of a plurality of data groups having different attributes based on first hash values ​​of each of the data of the plurality of data groups; a first acquisition process in which the second device acquires the plurality of data groups and a second hash value generated by the first generation process; an anonymization process in which the second device generates a first hash value of each data of a data group having a specific attribute among the plurality of data groups in response to an anonymization request from the third device; a second acquisition process in which, as a result of the third device transmitting an anonymization request for the data group having the specific attribute to the second device, the third device acquires from the second device a first hash value of each data of the data group having the specific attribute, a data group having remaining attributes among the plurality of data groups excluding the data group having the specific attribute, and the second hash value; a second generation process in which the third device generates a first hash value for each piece of data in the data group having the remaining attribute acquired by the second acquisition process, and generates a third hash value based on the first hash value for each piece of data in the data group having the remaining attribute and the first hash value for each piece of data in the data group having the specific attribute acquired by the second acquisition process; a verification process in which the third device verifies the validity of the anonymization process by determining whether or not the second hash value acquired by the second acquisition process matches the third hash value generated by the second generation process; In the first generation process, the first device generates a first hash value of the data in accordance with a processing rule corresponding to the attribute, and generates second hash values ​​of the plurality of data groups based on the first hash value of the data; In the first acquisition process, the second device acquires the processing rule from the first device, In the anonymization process, the second device generates a first hash value of each data of the data group having the specific attribute in accordance with the processing rule; In the second acquisition process, the third device acquires the processing rule from the second device, In the second generation process, the third device generates a first hash value of each data of the data group of the remaining attributes in accordance with the processing rule. An anonymization system comprising:

2. An anonymization system as claimed in claim 1, the first device has a random number data table that holds random number data, which is a combination of the data and a random value, for each of the data; In the first generation process, when the processing rule is to tokenize the data, the first device generates a first hash value of each data of the data group having a tokenization target attribute based on a data group having a tokenization target attribute to which the tokenization is applied and a random value of each data of the data group; In the first acquisition process, the second device acquires the random number data for each of the data from the first device and generates the random number data table; In the anonymization process, when the specific attribute is the attribute to be pseudonymized, the second device generates a first hash value of each data of the data group of the specific attribute based on each data of the data group of the specific attribute and a random value thereof. An anonymization system comprising:

3. An anonymization system as claimed in claim 1, the first device has a random number data table that holds random number data, which is a combination of the data and a random value thereof, for each of the data; In the first generation process, when the processing rule is to delete the data, the first device generates a first hash value of each data of the data group having an attribute to be deleted based on each data of the data group having an attribute to be deleted to which the deletion is applied, and its random value and unique value; In the first acquisition process, the second device acquires the random number data for each of the data from the first device and generates the random number data table; In the anonymization process, when the specific attribute is the attribute to be deleted, the second device generates a first hash value of each data of the data group of the specific attribute based on each data of the data group of the specific attribute, its random value, and the unique value. An anonymization system comprising:

4. An anonymization system as claimed in claim 3, In the anonymization process, when the anonymization request is for a portion of data of the specific attribute, the second device generates a first hash value of the portion of data of the specific attribute; In the second acquisition process, the third device acquires, as a result of transmitting the anonymization request to the second device, a first hash value of the part of data of the specific attribute from the second device, a remaining data group excluding the part of data of the specific attribute from the multiple data groups, and the second hash value, acquires verification random number data which is a combination of other data of the specific attribute other than the part of data of the specific attribute and its random value, and generates a verification random number data table; In the second generation process, the third device generates a first hash value for each data of the data group having attributes other than the specific attribute among the remaining data group acquired by the second acquisition process, generates a first hash value for each data of the data group having the other attributes among the remaining data group acquired by the second acquisition process, generates a first hash value for the data of the specific attribute among the remaining data group acquired by the second acquisition process based on the data, its random number value, and a unique value if the data exists in the verification random number data table, and generates the third hash value based on the first hash value of each data of the remaining data group and the first hash value of a part of the data of the specific attribute. An anonymization system comprising:

5. An anonymization system as claimed in claim 1, The first device has a random number data table that holds random number data, which is a combination of the data and a random value thereof, for each of the data, and a generalization rule table that specifies combinations of data before and after generalization, In the first generation process, when the processing rule is generalization of the data, the first device generates a first hash value of each data of the data group of the attribute to be generalized based on the data of the attribute to be generalized to which the generalization is applied, its random value and unique value, and the number of conversions from before generalization to after generalization according to the generalization rule table; In the first acquisition process, the second device acquires the random number data for each of the data from the first device to generate the random number data table, acquires a combination of data before and after generalization of the generalization rule table to generate the generalization rule table, In the anonymization process, when the specific attribute is the attribute to be generalized, the second device generates a first hash value for each of the data group of the specific attribute based on each data of the data group of the specific attribute, its random value and unique value, and the number of conversions. An anonymization system comprising:

6. An anonymization system having a first device, a second device capable of communicating with the first device, and a third device capable of communicating with the second device, a first generation process in which the first device generates second hash values ​​of a plurality of data groups having different attributes based on first hash values ​​of each of the data of the plurality of data groups; a first acquisition process in which the second device acquires the plurality of data groups and a second hash value generated by the first generation process; an anonymization process in which the second device generates a first hash value of each data of a data group having a specific attribute among the plurality of data groups in response to an anonymization request from the third device; a second acquisition process in which, as a result of the third device transmitting an anonymization request for the data group having the specific attribute to the second device, the third device acquires from the second device a first hash value of each data of the data group having the specific attribute, a data group having remaining attributes among the plurality of data groups excluding the data group having the specific attribute, and the second hash value; a second generation process in which the third device generates a first hash value for each piece of data in the data group having the remaining attribute acquired by the second acquisition process, and generates a third hash value based on the first hash value for each piece of data in the data group having the remaining attribute and the first hash value for each piece of data in the data group having the specific attribute acquired by the second acquisition process; a verification process in which the third device verifies the validity of the anonymization process by determining whether or not the second hash value acquired by the second acquisition process matches the third hash value generated by the second generation process; In the anonymization process, the second device generates a first hash value of a part of data of the specific attribute for which anonymization is requested, In the second acquisition process, the third device acquires, as a result of transmitting the anonymization request for the part of the data of the specific attribute to the second device, a first hash value of the part of the data of the specific attribute from the second device, a remaining data group excluding the part of the data of the specific attribute from the plurality of data groups, and the second hash value; In the second generation process, the third device generates a first hash value of each data of the remaining data group acquired by the second acquisition process, and generates the third hash value based on the first hash value of each data of the remaining data group having an attribute and a first hash value of a part of the data of the specific attribute acquired by the second acquisition process. An anonymization system comprising:

7. An anonymization method using an anonymization system having a first device, a second device capable of communicating with the first device, and a third device capable of communicating with the second device, a first generation process in which the first device generates second hash values ​​of a plurality of data groups having different attributes based on first hash values ​​of each of the data of the plurality of data groups; a first acquisition process in which the second device acquires the plurality of data groups and a second hash value generated by the first generation process; an anonymization process in which the second device generates a first hash value of each data of a data group having a specific attribute among the plurality of data groups in response to an anonymization request from the third device; a second acquisition process in which, as a result of the third device transmitting an anonymization request for the data group having the specific attribute to the second device, the third device acquires from the second device a first hash value of each data of the data group having the specific attribute, a data group having remaining attributes among the plurality of data groups excluding the data group having the specific attribute, and the second hash value; a second generation process in which the third device generates a first hash value for each piece of data in the data group having the remaining attribute acquired by the second acquisition process, and generates a third hash value based on the first hash value for each piece of data in the data group having the remaining attribute and the first hash value for each piece of data in the data group having the specific attribute acquired by the second acquisition process; a verification process in which the third device verifies the validity of the anonymization process by determining whether or not the second hash value acquired by the second acquisition process matches the third hash value generated by the second generation process; In the first generation process, the first device generates a first hash value of the data in accordance with a processing rule corresponding to the attribute, and generates second hash values ​​of the plurality of data groups based on the first hash value of the data; In the first acquisition process, the second device acquires the processing rule from the first device, In the anonymization process, the second device generates a first hash value of each data of the data group having the specific attribute in accordance with the processing rule; In the second acquisition process, the third device acquires the processing rule from the second device, In the second generation process, the third device generates a first hash value of each data of the data group of the remaining attributes in accordance with the processing rule.

13. An anonymization method comprising:

8. The anonymization method according to claim 7, the first device has a random number data table that holds random number data, which is a combination of the data and a random value, for each of the data; In the first generation process, when the processing rule is to tokenize the data, the first device generates a first hash value of each data of the data group having a tokenization target attribute based on a data group having a tokenization target attribute to which the tokenization is applied and a random value of each data of the data group; In the first acquisition process, the second device acquires the random number data for each of the data from the first device and generates the random number data table; In the anonymization process, when the specific attribute is the attribute to be pseudonymized, the second device generates a first hash value of each data of the data group of the specific attribute based on each data of the data group of the specific attribute and a random value thereof.

13. An anonymization method comprising:

9. The anonymization method according to claim 7, the first device has a random number data table that holds random number data, which is a combination of the data and a random value thereof, for each of the data; In the first generation process, when the processing rule is to delete the data, the first device generates a first hash value of each data of the data group having an attribute to be deleted based on each data of the data group having an attribute to be deleted to which the deletion is applied, and its random value and unique value; In the first acquisition process, the second device acquires the random number data for each of the data from the first device and generates the random number data table; In the anonymization process, when the specific attribute is the attribute to be deleted, the second device generates a first hash value of each data of the data group of the specific attribute based on each data of the data group of the specific attribute, its random value, and the unique value.

13. An anonymization method comprising:

10. The anonymization method according to claim 9, In the anonymization process, when the anonymization request is for a portion of data of the specific attribute, the second device generates a first hash value of the portion of data of the specific attribute; In the second acquisition process, the third device acquires, as a result of transmitting the anonymization request to the second device, a first hash value of the part of data of the specific attribute from the second device, a remaining data group excluding the part of data of the specific attribute from the multiple data groups, and the second hash value, acquires verification random number data which is a combination of other data of the specific attribute other than the part of data of the specific attribute and its random value, and generates a verification random number data table; In the second generation process, the third device generates a first hash value for each data of the data group having attributes other than the specific attribute among the remaining data group acquired by the second acquisition process, generates a first hash value for each data of the data group having the other attributes among the remaining data group acquired by the second acquisition process, generates a first hash value for the data of the specific attribute among the remaining data group acquired by the second acquisition process based on the data, its random number value, and a unique value if the data exists in the verification random number data table, and generates the third hash value based on the first hash value of each data of the remaining data group and the first hash value of a part of the data of the specific attribute.

13. An anonymization method comprising:

11. The anonymization method according to claim 7, The first device has a random number data table that holds random number data, which is a combination of the data and a random value thereof, for each of the data, and a generalization rule table that specifies combinations of data before and after generalization, In the first generation process, when the processing rule is generalization of the data, the first device generates a first hash value of each data of the data group of the attribute to be generalized based on the data of the attribute to be generalized to which the generalization is applied, its random value and unique value, and the number of conversions from before generalization to after generalization according to the generalization rule table; In the first acquisition process, the second device acquires the random number data for each of the data from the first device to generate the random number data table, acquires a combination of data before and after generalization of the generalization rule table to generate the generalization rule table, In the anonymization process, when the specific attribute is the attribute to be generalized, the second device generates a first hash value for each of the data group of the specific attribute based on each data of the data group of the specific attribute, its random value and unique value, and the number of conversions.

13. An anonymization method comprising:

12. An anonymization method using an anonymization system having a first device, a second device capable of communicating with the first device, and a third device capable of communicating with the second device, a first generation process in which the first device generates second hash values ​​of a plurality of data groups having different attributes based on first hash values ​​of each of the data of the plurality of data groups; a first acquisition process in which the second device acquires the plurality of data groups and a second hash value generated by the first generation process; an anonymization process in which the second device generates a first hash value of each data of a data group having a specific attribute among the plurality of data groups in response to an anonymization request from the third device; a second acquisition process in which, as a result of the third device transmitting an anonymization request for the data group having the specific attribute to the second device, the third device acquires from the second device a first hash value of each data of the data group having the specific attribute, a data group having remaining attributes among the plurality of data groups excluding the data group having the specific attribute, and the second hash value; a second generation process in which the third device generates a first hash value for each piece of data in the data group having the remaining attribute acquired by the second acquisition process, and generates a third hash value based on the first hash value for each piece of data in the data group having the remaining attribute and the first hash value for each piece of data in the data group having the specific attribute acquired by the second acquisition process; a verification process in which the third device verifies the validity of the anonymization process by determining whether or not the second hash value acquired by the second acquisition process matches the third hash value generated by the second generation process; In the anonymization process, the second device generates a first hash value of a part of data of the specific attribute for which anonymization is requested, In the second acquisition process, the third device acquires, as a result of transmitting the anonymization request for the part of the data of the specific attribute to the second device, a first hash value of the part of the data of the specific attribute from the second device, a remaining data group excluding the part of the data of the specific attribute from the plurality of data groups, and the second hash value; In the second generation process, the third device generates a first hash value of each data of the remaining data group acquired by the second acquisition process, and generates the third hash value based on the first hash value of each data of the remaining data group having an attribute and a first hash value of a part of the data of the specific attribute acquired by the second acquisition process.

13. An anonymization method comprising:

Citation Information

Patent Citations

  • Personal information browsing / update system and method

    JP2006285490A

  • Anonymization system and anonymization method

    JP2020077256A

  • Information management system, anonymizing method, and storage medium

    WO2008069011A1