JP6685968B2 - Data management system, data management method, and data management program - Google Patents

Description

  The present invention relates to a system, method and program for managing personal information.

2. Description of the Related Art Heretofore, a method of selectively disclosing personal information has been studied. For example, Non-Patent Document 1 proposes a privacy policy manager as a framework for more easily interpreting a privacy policy.
Further, for example, Non-Patent Documents 2 and 3 present a system for storing verifiable information that is anonymized by using a block chain.

Kiyomoto, S.K. , Nakamura, T .; , Takasaki, H .; Watanabe, R .; , Miyake, Y .; : PPM: Privacy policy manager for personalized services. In: Security Engineering and Intelligence Informatics. Springer (2013) 377-392. KPMG launches Digital Ledger Services to help companies implement blockchain technology: Internet <URL: https://home.kpmg.com/sg/en/home/media/press-releases/2016/11-kpmg-ledgerunch services-to-help-companies-implement-blockchain-technology.html> Deloitte SmartID: Internet <URL: http://www.deloitte.co.uk/smartid/>

However, blockchain is regarded as a so-called trustless system. That is, in a public blockchain including a large number of participants, there is a possibility of conspiracy or falsification of majority opinion, and even in a private blockchain including a limited number of participants, there is a risk of conspiracy.
Also, the public blockchain is not scalable as all participants maintain a very large hashchain. Furthermore, maintaining the information of all users in a single blockchain is resource-intensive, which is not efficient in some situations.

  An object of the present invention is to provide a data management system, a data management method, and a data management program that can efficiently limit and disclose the disclosure range based on user data dependency relationships.

  The data management system according to the present invention is dependent on a leaf node storing user data, a hash node generating unit that generates a hash node including a hash value of the leaf node and a digital signature, and first user data. A leaf node generation unit that stores a pointer to the hash node corresponding to the first user data in a new leaf node together with related second user data.

  The data management system includes an upper node generating unit that generates a multi-level tree structure by providing an upper node that includes a hash value of data that is obtained by connecting the hash values stored in the plurality of hash nodes and a digital signature. May be.

  The hash value may be calculated by a one-way hash function using a random number set for each user.

  A data management method according to the present invention, for a leaf node in which user data is stored, a hash node generating step of generating a hash node including a hash value of the leaf node and a digital signature; The computer executes a leaf node generation step of storing a pointer pointing to the hash node corresponding to the first user data in a new leaf node together with the related second user data.

  A data management program according to the present invention, for a leaf node storing user data, a hash node generating step of generating a hash node including a hash value of the leaf node and a digital signature; And a leaf node generation step of storing a pointer pointing to the hash node corresponding to the first user data in a new leaf node together with the related second user data.

  According to the present invention, the disclosure range can be efficiently limited and verified based on the dependency relationship of user data.

It is a block diagram showing functional composition of a data management system concerning an embodiment. It is a figure which shows the structure of VI graph which concerns on embodiment. It is a figure which shows the information stored in the node in each level of the VI graph which concerns on embodiment. It is a figure which shows the relationship between the leaf node and hash node which concern on embodiment. It is a sequence diagram which shows the addition operation to the empty VI graph which concerns on embodiment. It is a sequence diagram which shows the addition operation to the existing VI graph which concerns on embodiment. It is a figure which shows the verification process by the verifier organization which concerns on embodiment.

Hereinafter, an example of the embodiment of the present invention will be described.
The data management system 1 according to the present embodiment manages a VI (Verifiable Information) graph to selectively disclose user data and enable verification of these user data.

FIG. 1 is a block diagram showing a functional configuration of a data management system 1 according to this embodiment.
The data management system 1 is implemented by a plurality of information processing devices (computers). For example, in the terminal 2 of the user who is the data owner, the control unit 10 functions as the following units by executing the software (data management program) stored in the storage unit 20.
The control unit 10 includes a hash node generation unit 11, a leaf node generation unit 12, and an upper node generation unit 13.

The hash node generation unit 11 generates a hash node including a hash value of this leaf node and a digital signature for the leaf node in which the user data is stored.
That is, a hash node capable of verifying user data and a digital signature is generated above the leaf node.

The leaf node generation unit 12 stores the pointer pointing to the hash node corresponding to the first user data in the new leaf node together with the second user data having a dependency relationship with the first user data.
That is, the new leaf node includes a pointer that points to a hash node above the existing leaf node, and the hash value of the entire new leaf node including the user data and the pointer is stored in the new hash node.

The upper node generation unit 13 includes an upper node that includes a hash value of data obtained by concatenating hash values stored in a plurality of hash nodes and an electronic signature, and generates a multi-stage tree structure.
As a result, in the tree structure having the root node as the apex, the leaf node generates a VI graph in which the dependency pointer is added to the hash node.

Data of upper nodes of the generated VI graph excluding the leaf node in which the user data is stored is stored in the management server 3 of the metadata provider and provided to the verifier.
Here, the hash value stored in the node higher than the leaf node is calculated by a one-way hash function using a random number set for each user. These hash values are used to verify the requested user data and the corresponding digital signature.

FIG. 2 is a diagram showing the structure of the VI graph according to this embodiment.
In the VI graph, each branch in the tree structure can be independently verified, and the user data in the branch not requested by the verifier is hidden.
The VI graph has a depth of 4 levels (root level, group level, hash level, data level), for example. The actual user data is stored in the leaf node, which is the fourth level.

  The VI graph is for verifiability and selective information release, and has a tree structure more generalized than a binary hash tree or the like for high-speed search from the root node. That is, a non-leaf node can have more than two child branches (lower nodes).

There are four types of entities related to VI graph generation and management, and data processing is performed by an information processing device included in each entity.
(A) A user who is the owner of the data.
(B) An organizational entity that certifies user data, such as a Public Safety Commission that issues a driver's license.
(C) A metadata provider that holds at least a part of the VI graph and provides it for verification.
(D) A verifier organization such as a service provider that verifies information managed by the VI graph.

  Each user has a personal VI graph that manages his or her user data, some of which are managed by one or more metadata providers. Note that depending on the type of user data, it may not be approved by the relevant information issuing authority, and in this case, the organizational entity is unnecessary.

Here, there is a dependency relationship (pointer) between the hash level and the data level. In the example of FIG. 2, one-to-one, one-to-many, and many-to-one dependency relationships are shown.
This pointer is provided for a past hash node when a new leaf node is created. That is, in the time series from left to right illustrated in FIG. 2, a pointer is provided from the leaf node to the hash node to the left of itself.

  FIG. 3 is a diagram showing information stored in nodes at each level of the VI graph according to this embodiment.

[Data level]
The leaf nodes L0, ..., L9 include individual user data L_i such as national identification information, an address, and a driver's license. Furthermore, each leaf node may include a pointer, which will be described later, indicating a dependency relationship.

[Hash level]
The hash nodes D0, ..., D9 include the hash value h (L_i, r_u) calculated by the cryptographically secure one-way hash function h for the user data L_i. Here, r_u is a random value different for each user. As a result, even if the same data is used, the hash value will be different if the user is different. Therefore, since there is no relation between the hash values, the user data is hidden by analogy with the hash values.

  Each hash node includes, in addition to the hash value of the corresponding user data, the electronic signature sig_u (h (L_i, r_u)) of the user for this hash value. The hash node may further include the signature sig_o (h (L_i, r_u) of the organizational entity corresponding to the user data. For example, if leaf node L5 represents the current address of the user, hash node D5 may include L5. , The user's digital signature, and the digital signature of the municipality as an organizational entity.

[Group level]
The group nodes G0, ..., G2 are defined by the user and are provided for the purpose of dividing an information group for semantic grouping according to data type or privacy protection at the time of verification. Each group node contains a hash value of data obtained by concatenating the hash values stored in the child nodes immediately below. In addition, each group node also includes the electronic signature of the user.

[Root level]
Similar to the group node, the root node R includes a hash value of data obtained by concatenating the hash values stored in the child nodes immediately below. In addition, the root node includes a random value r_u unique to the user, the electronic signature of the user, and the electronic signature of the metadata provider that maintains and manages the VI graph.

FIG. 4 is a diagram showing the relationship between leaf nodes and hash nodes according to this embodiment.
The dependency between user data is defined by a unidirectional pointer from a data level leaf node to a hash level hash node.
One or more pointers may be stored in the leaf node together with the user data, and each pointer points to a past hash node (Dx and Dy). The hash node D above the leaf node stores the entire hash value including these pointers and user data.

The dependency indicated by the pointer is used to retrieve the related data. The pointer is defined when adding new user data to the VI graph and does not change thereafter. For example, if a user with nationality of country C1 obtains a work permit in country C2 and resides in C2, the leaf node is accompanied by a pointer to a hash node signed for citizenship, along with the work permit data. Is included. This indicates a semantic dependency where the visa depends on a particular nationality.
Further, for example, the leaf node indicating the current address after relocation includes a pointer to the hash node indicating the address before relocation, and the past address information of the user is sequentially searched as needed.

  The definition of the dependency is at the user's discretion, but the organizational entity that is the signing authority of the corresponding hash node may reject it if it determines that the dependency specified by the user is irrelevant, or vice versa. The user may be requested to set the dependency relationship of. The dependency may be any of one-to-one, one-to-many, many-to-one, and many-to-many, and further may span groups. All dependencies are represented by pointers to the past.

Next, a method of maintaining and managing the VI graph in the data management system 1 will be described.
Some of the VI graphs are stored by the user as well as the metadata provider. When there are multiple metadata providers, the VI graph may be stored in the multiple metadata providers and Erase Coding may be performed to ensure reliability.
The user saves all nodes except the root node and the metadata provider saves all nodes except leaf nodes. This allows the verifier organization to verify some or all of the VI graph stored by the metadata provider without the user having to publish a copy of highly confidential user data, such as personal information.

[Additional operation]
The add operation allows the user to add a data level leaf node to an empty graph or an existing graph. In either case, it is assumed that the user has already generated the public key / private key pair.

FIG. 5 is a sequence diagram showing an adding operation to an empty VI graph according to this embodiment.
In step S1 (initialization step), the user generates a random value as an initial process.
In step S2 (data generation step), the user calculates a hash value of user data and adds a digital signature.

  In step S3 (signature acquisition step), the user requests the organizational entity for a signature and acquires the electronic signature for the leaf node.

  In step S4 (graph generation step), the user calculates a hash value stored in the group node and adds a digital signature. Further, the user calculates a hash value stored in the root node and adds a digital signature. The user also publishes the random value to the metadata provider.

  In step S5 (publication step), the user requests the metadata provider for a signature and obtains an electronic signature for the root node. After that, the user publishes the generated VI graph to the metadata provider.

FIG. 6 is a sequence diagram showing an adding operation to an existing VI graph according to this embodiment.
In step S11 (data generation step), the user defines the dependency relationship of the user data and generates a leaf node. Further, the user calculates the hash value of the leaf node and adds the digital signature.

  In step S12 (signature acquisition step), the user requests the organizational entity for a signature and acquires the electronic signature for the leaf node.

  Hereinafter, steps S13a to S14a or steps S13b to S14b are executed depending on whether the leaf node is added to a new group or an existing group.

  In step S13a (group generation step), the user calculates a hash value stored in the group node and adds a digital signature. Further, the user updates the hash value stored in the root node.

  In step S14a (verification step), the metadata provider verifies the updated hash value of the root node and adds an electronic signature to the root node.

  In step S13b (group updating step), the user updates the hash value stored in the group node and adds a digital signature. Further, the user updates the hash value stored in the root node.

  In step S14b (verification step), the metadata provider verifies the updated hash value of the group node and the hash value of the root node, and adds an electronic signature to the root node.

  In step S15 (publication step), the user acquires an electronic signature for the root node from the metadata provider. The user then publishes the updated VI graph to the metadata provider.

[Delete operation]
The user can delete only the leaf node from the local storage by the delete operation. As a result, the corresponding hash node is not deleted, but the original user data cannot be restored. The remaining hash node is used for verification of other nodes without disclosing the corresponding user data.
Here, deleting the electronic signature by the organizational entity during the verification process also prevents the verifier from inferring the original user data from the hash value alone.

[Move operation]
The user can move the leaf node and the corresponding hash node from one group to another group by the move operation. The move operation is the delete operation followed by the add operation, but since the moving node is not deleted from the graph, a dependency between groups may occur.

  In order to delete the existing leaf node L_k from the lower order of the group node G_k, the user has to connect the hash value of the hash value L_k included in G_k and the hash value included in another hash node to the hash value of the data. Must be proved to the metadata provider. When this is verified, the hash value of G_k is updated to the hash value for the data obtained by removing the hash value of L_k from the concatenated data, and the electronic signature is added by the user.

  After that, the move operation is completed by the add operation to the graph, but since the moving hash node already has the necessary digital signature, the signature acquisition step among the add operations to the existing graph described above. Is omitted.

FIG. 7 is a diagram showing a verification process by the verifier organization according to the present embodiment.
First, a verifier organization such as a service provider acquires the following two data sets from each of the user and the metadata provider.

(From user)
・ Required user data to be verified ・ Hash node with digital signature corresponding to the required user data ・ Hash node belonging to the same group as the hash node corresponding to the required user data ・ Digital signature with no required user data under it Group node

(From metadata provider)
-A random value unique to the user-A hash node with a digital signature corresponding to the requested user data-A group node with a digital signature above the requested user data-A root node with a digital signature

  In step S21, the verifier organization calculates a hash value of the requested user data provided by the user, determines whether or not this hash value matches that obtained from both the user and the metadata provider, and Check whether the digital signature is correct. The verifier organization also confirms the dependency relationship at this stage.

  In step S22, the verifier organization concatenates the information of all the other hash nodes provided by the user, then calculates the hash value corresponding to the group node, and the hash value and the electronic signature are obtained from the metadata provider. Check if it matches the one obtained.

  In step S23, the verifier organization calculates the hash value corresponding to the root node after concatenating the information of all other group nodes provided by the user, and obtains this hash value and the electronic signature from the metadata provider. Check whether it matches the one

  Here, in order to verify the user data at the data level, it is necessary for the user to provide other hash level information of the same group. Therefore, in order to improve the efficiency of the verification process, it is desirable that the number of hash nodes belonging to the group is small, that is, the number of groups is large.

  According to this embodiment, the data management system 1 manages a VI graph including a dependency relationship between a leaf node and a hash node. Therefore, the data management system 1 efficiently limits the disclosure range based on the user data dependency while suppressing the storage capacity and enabling high-speed processing as compared with the method based on the block chain. Can be verified.

  Further, the data management system 1 is provided with a group node and a root node and constructs a multi-stage tree structure to form a data group having a semantic unity, and selectively discloses and verifies user data. Can be made more efficient.

  Further, the data management system 1 can increase the confidentiality of the user data by changing the hash value among the users even for the same user data by the hash function using the random value unique to the user.

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments. In addition, the effects described in the present embodiment are merely enumeration of the most suitable effects resulting from the present invention, and the effects according to the present invention are not limited to those described in the present embodiment.

  The data management method by the data management system 1 is realized by software. When implemented by software, a program forming the software is installed in an information processing device (computer). Further, these programs may be recorded on a removable medium such as a CD-ROM and distributed to users, or may be distributed by being downloaded to a user's computer via a network. Further, these programs may be provided to the user's computer as a Web service via the network without being downloaded.

1 Data Management System 2 Terminal 3 Management Server 10 Control Unit 11 Hash Node Generation Unit 12 Leaf Node Generation Unit 13 Upper Node Generation Unit 20 Storage Unit

Claims (5)

For a leaf node in which user data is stored, a hash node generation unit that generates a hash node including the hash value and electronic signature of the leaf node,
A leaf node generation unit that stores, in a new leaf node, a pointer that points to the hash node corresponding to the first user data, together with second user data that has a dependency relationship with the first user data. system.   The upper node generation unit that generates a multi-level tree structure by providing an upper node that includes a hash value of data that is obtained by concatenating the hash values stored in a plurality of the hash nodes and an electronic signature. Data management system.   The data management system according to claim 1 or 2, wherein the hash value is calculated by a one-way hash function using a random number set for each user. For a leaf node in which user data is stored, a hash node generating step of generating a hash node including a hash value of the leaf node and a digital signature,
The computer executes a leaf node generation step of storing, in a new leaf node, a pointer pointing to the hash node corresponding to the first user data, together with the second user data having a dependency relationship with the first user data. Data management method. For a leaf node in which user data is stored, a hash node generating step of generating a hash node including a hash value of the leaf node and a digital signature,
A leaf node generating step of storing, in a new leaf node, a pointer pointing to the hash node corresponding to the first user data, together with the second user data having a dependency relationship with the first user data; Data management program for

Images