Information processing system, information processing method, and computer program

Abstract

The objective is to provide an information processing system that can regenerate the data structure necessary for business continuity even if the contents of a file are lost, while also efficiently storing structural information and ensuring information security. [Solution] The system includes a structure extraction means for extracting structure information independent of the file contents, a mapping generation means for removing business-meaning identifiers or mapping them to a generalized non-business-meaning representation while maintaining the topology of the structure information, a storage means for storing the non-business-meaning representation in a storage area separate from the area where the file is stored, and a reconstruction means for deriving structure information from the non-business-meaning representation and regenerating the data structure.

Description

[Technical Field]

[0001] The present invention relates to an information processing system, an information processing method, and a computer program, and particularly relates to a technique for reconstructing a data structure when a file content loss event occurs. [Background Art]

[0002] In recent years, corporate business systems are operated by storing structured data on various storage media such as databases, file servers, and cloud storage. These pieces of data are accompanied by structural information such as reference relationships between tables, constraint conditions, index structures, and metadata. If the structural information is lost, it may be difficult to resume business processing even if the content of the data physically remains.

[0003] Conventional backup technology is mainly based on a method of copying and storing the file content itself, but this method has several problems. For example, if backup data is under the same management system, there is a possibility that it will be lost at the same time. In addition, complete replication of large-capacity data requires a large amount of storage capacity and communication bandwidth. Furthermore, restoration from backup is often premised on full restoration, making it difficult to reconstruct only a part of the structure.

[0004] In addition, conventional access control technologies and encryption technologies mainly aim to prevent unauthorized use of data or control access, and do not sufficiently address the problem of reconstructing the data structure itself after file content is lost. These technologies are premised on retaining or protecting data content, and do not have the main objective of extracting and reconstructing only structural information. [Prior Art Documents] [Patent Documents]

[0005] [Patent Document 1] Japanese National Publication of International Patent Application No. 2013-529799 [Patent Document 2] Japanese Patent Publication No. 2014-153761 [Overview of the project] [Problems that the invention aims to solve]

[0006] Conventional backup technologies, such as those disclosed in Patent Document 1, primarily aim to save or restore file contents, and do not have the technical core objective of regenerating only the data structure after file contents have been lost. Furthermore, access control technologies, such as those disclosed in Patent Document 2, aim to protect data contents or control access, and therefore have different objectives from the reconstruction of structural information.

[0007] The present invention aims to provide an information processing system that can regenerate the data structure necessary for business continuity even if the contents of a file are lost, and that can efficiently store only the structural information while ensuring information security. [Means for solving the problem]

[0008] To solve the above problems, the information processing system according to the present invention is: An information processing system that reconstructs the data structure after the loss of file contents, A structural extraction means that extracts structural information independent of the file's contents, A mapping generation means that, while maintaining the topology of the structural information, removes business-meaning identifiers or performs a mapping transformation to a generalized non-business-meaning representation, Storage means for storing the non-business semantic expression in a storage area that is logically or physically separated from the area where the file is stored, A reconstruction means for deriving structural information from the aforementioned non-business semantic expression and regenerating the data structure based on the derived structural information, An information processing system characterized by comprising the following features. [Effects of the Invention]

[0009] According to the present invention, even if the contents of a file are lost, it is possible to reconstruct the data structure necessary for business continuity based on structural information derived from non-business semantic representations that have been stored in advance. Furthermore, since the non-business semantic representations are stored in a format in which identifiers with business meaning have been removed or generalized, even if such information is leaked, it is possible to suppress the direct reproduction of business content.

[0010] Furthermore, by performing a mapping transformation that removes business meaning while preserving the structural topology, structural recovery can be achieved that does not depend on saving or encrypting the file contents. In addition, by combining consistency control based on domain-specific identification information and an identification pattern superimposition configuration on the display output, a multi-layered protection can be achieved that prevents reconstruction in unauthorized environments and enables leakage tracking.

[0011] Furthermore, since only structural topology information is stored and no actual business data content is stored, the amount of data to be stored can be reduced, and the risk of business information leakage can be lowered. Moreover, since the present invention can regenerate the data structure based on structural topology information without storing the business data content itself, it enables data content-independent regeneration. Therefore, even if business data content is encrypted, damaged, or lost due to ransomware or other events, the data structure necessary for business continuity can be reconstructed. In addition, since personal information, transaction information, and other business data content are not stored, the risk of information leakage caused by the storage of content information can be reduced. [Brief explanation of the drawing]

[0012] [Figure 1] This is an overall configuration diagram of the information processing system according to the present invention. [Figure 2] This is a functional block diagram of the structural mapping reconstruction core according to the present invention. [Figure 3] This is a flowchart of the information processing system according to the present invention. [Figure 4] This figure shows an example of the concept of mapping transformation according to the present invention. [Figure 5] It is a diagram illustrating the concept of region-specific identification information and matching determination according to the present invention. [Figure 6] It is a diagram illustrating the concept of superimposition and analysis of identification patterns in display output according to the present invention. [Figure 7A] It is a diagram showing an example of transition of a graph structure in mapping conversion according to the present invention. [Figure 7B] It is a diagram showing an example of a mapping conversion element according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The following embodiments are merely illustrative and do not limit the present invention in any way. Therefore, the present invention can naturally be subjected to various improvements and modifications without departing from the scope of the gist thereof.

[0014] Figure 1 is an overall configuration diagram of an information processing system according to the present embodiment. The present system includes a business file system 10, a structural mapping reconstruction core 20, a restored file system 18, a control means 16, and a display control means 19, which are configured to be connected to each other.

[0015] The business file system 10 is a file system that stores business data to be protected. The structural mapping reconstruction core 20 is a core processing unit that performs structure extraction, mapping conversion, storage, and reconstruction processing on the business file system 10. The restored file system 18 is a file system having a data structure regenerated by the reconstruction processing.

[0016] The control means 16 is a control unit that controls whether the structural mapping reconstruction core 20 can be executed, and performs matching determination based on region-specific identification information at the time of generation. The display control means 19 is an output control unit that superimposes an identification pattern on the display output of the restored file system 18.

[0017] FIG. 2 is a functional block configuration diagram of a structural mapping reconstruction core 20 included in an information processing system. The structural mapping reconstruction core 20 is configured to include a structure extraction means 11, a mapping generation means 12, a storage means 13, and a reconstruction means 14. The reconstruction means 14 can be configured to internally include a consistency verification means 15.

[0018] The structure extraction means 11 extracts structure information independent of file contents from each file in a business file system 10. Here, the structure information refers to information including table definitions, column configurations, primary key and foreign key relationships, constraint conditions, index definitions, data type definitions, and their mutual relationships (topology). The structure extraction means 11 can be configured to be automatically executed when a business file is created or updated.

[0019] The present invention does not target the business data content itself for processing, but targets information processing that extracts structure information accompanying the business data, models the structure information, performs mapping conversion and other structural operations on the structure information, and retains the information in a regenerable form.

[0020] The mapping generation means 12 maps and converts the extracted structure information into a non-business semantic expression 32 obtained by removing or generalizing identifiers having business meanings while retaining the topology of the structure. Here, the "identifiers having business meanings" refer to table names, column names, index names, field names, and other names, symbols or character strings that directly or indirectly indicate business contents, business functions or business objects. In addition, the "non-business semantic expression" refers to an expression obtained by removing, generalizing, re-encoding or abstracting the business meaning of the aforementioned identifiers, which is an expression that cannot immediately understand the original business content when standing alone, but retains information related to structural topology.

[0021] The storage means 13 stores the non-business semantic expression 32 in a storage area that is logically or physically separated from the original business file system 10. The storage means 13 can be configured such that it does not store the contents of business files (such as row data), but stores the non-business semantic expression (and metadata accompanying it).

[0022] The reconstruction means 14 derives structural information from the saved non-business semantic representations 32 and regenerates a data structure equivalent to the original structure. Here, "data structure equivalent to the original structure" does not mean restoring the content of the original business data, but rather refers to a data structure in which primary key relationships, foreign key relationships, constraints, index structures, attribute configurations, and other structural relationships substantially correspond. The data structure regenerated here does not restore the content of the original business data, but rather regenerates an equivalent structure based on the preserved structural relationships.

[0023] Since the non-business semantic representation 32 includes information about the structural topology 31, even if a part of it is missing or corrupted, the data structure can be reconstructed to the extent that it can be regenerated based on the remaining topology information.

[0024] The consistency verification means 15 verifies the consistency of the regenerated data structure, including primary key uniqueness, foreign key referential integrity, constraints, and index structure. Furthermore, even if foreign key constraints include multiple reference relationships or dependencies are chained due to cascading deletion constraints, the consistency verification means 15 can obtain a consistently regenerated data structure by verifying the consistency of referential integrity and constraints.

[0025] Figure 3 is a flowchart of the data structure reconstruction process according to this embodiment.

[0026] First, in the file generation or update step S10, a business file is generated or updated. Next, in the structure extraction step S11, the structure extraction means 11 extracts structure information. In the mapping transformation step S12, the mapping generation means 12 generates a non-business semantic representation 32. In the saving step S13, the saving means 13 saves the non-business semantic representation 32 to a separate storage area.

[0027] In the file content loss detection step S14, the loss or corruption of file content is detected. Subsequently, in the area-specific identification information acquisition step S15, the area-specific identification information 17 assigned at the time of generation is acquired. Then, in the consistency determination step S16, the control means 16 determines the consistency with the identification information. If consistency is achieved, the derivation and regeneration step S17 is executed, and the reconstruction means 14 derives structural information from the saved non-business semantic representation 32 and regenerates the data structure. Then, in the consistency verification step S18, the consistency verification means 15 confirms the consistency of the regenerated structure. After confirmation, in the structure application step S19, the regenerated data structure is applied to the restored file system 18.

[0028] If consistency is not achieved, step S20 (rejection, logging, and re-authentication request) is executed, and the rebuild process is stopped.

[0029] After the reconstruction is complete, in the watermark superposition step S21, the display control means 19 superimposes the identification pattern onto the display output.

[0030] Figure 4 is a diagram illustrating an example of the concept of mapping transformation according to the present invention, showing the relationship between the original structural information 30, the structural topology 31, and the non-business semantic representation 32.

[0031] The original structural information 30 shows an example of a data structure defined in a business system. In this example, there are multiple tables, such as an Orders table and a Customers table, and each table has a primary key (PK), foreign keys (FK), multiple columns, constraints, and an index structure. At this stage, identifiers such as table names and column names directly contain business meaning.

[0032] The mapping generation means 12 extracts structural relationships from the structural information, representing tables as nodes (T1, T2, etc.) and foreign key reference relationships as edges, as shown in Figure 4, for example, and abstracts these as a structural topology 31. For example, the order table is represented as node T1 with 5 columns, and the customer table is represented as node T2 with 3 columns. In addition, the reference relationship between the foreign key of the order table and the primary key of the customer table is represented as edge T1.FK→T2.PK. The structural topology 31 includes information representing structural features such as the number of nodes, connection relationships, constraint types, and index structure, and the structure's framework is maintained by the nodes and edges.

[0033] Next, the mapping generation means 12 maps the structural topology 31 to a non-business semantic representation 32 in which business-meaning identifiers are removed or generalized. In the non-business semantic representation 32, business-meaning identifiers such as table names and column names are removed or generalized and replaced with encoded identifiers, random identifiers, hash values, or abstract symbols. On the other hand, structural relationships such as the number of nodes, connection relationships, constraint types, and index structure are preserved. For example, in the example in Figure 4, node T1 is represented as "x7f2a:[5,{r:3}]" which represents structural attributes including the number of columns and the number of connection relationships, node T2 is represented as an encoded identifier such as "k9e1b:[3,{r:1}]", and the reference relationship between nodes is represented as "link:x7f2a.2→k9e1b.0".

[0034] Thus, the mapping transformation in this invention aims to preserve structural relationships rather than the "content" that retains business meaning. Therefore, the generated non-business meaning representation 32 cannot reproduce business processing content on its own, but it enables structure derivation and regeneration by the reconstruction means 14 based on the preserved structural topology 31.

[0035] Furthermore, the mapping transformation and regeneration in this invention do not involve the transformation, transfer, integration, or rearrangement of the data content itself, but rather process the relationships between structural elements and the topology as the target of preservation.

[0036] The said mapping transformation may include multidimensional mappings, nonlinear transformations, coding processes, or random number generation processes, but is not limited to a specific transformation method.

[0037] For example, the mapping transformation can be configured to generate a non-business semantic representation 32 based on multiple feature quantities that constitute the structural information. Here, the multiple feature quantities can include the number of nodes, the number of attributes of each node, the number of edges, connection relationships, constraint types, index structure, reference direction, and other structural features.

[0038] The mapping transformation is not limited to simply replacing each feature with a one-to-one correspondence, but may also be configured to generate a non-business semantic representation 32 by a mapping operation that combines multiple features. The mapping operation is not limited to linear combinations, but may include non-linear operations such as product terms between features, thresholding, conditional branching, phase shifting, random number generation, and others.

[0039] Furthermore, the mapping transformation can be configured to be performed based on predetermined mapping rules. Here, "mapping rules" refers to correspondence relationships, calculation conditions, or transformation procedures for converting structural information to non-business semantic representations 32 and deriving structural information from said non-business semantic representations 32. The reconstruction means 14 derives structural information from the non-business semantic representations 32 in accordance with the mapping rules. The mapping rules may be stored separately from the storage means 13, or they may be stored in association with the non-business semantic representations 32.

[0040] This clarifies that the present invention does not restore content like cryptography or decryption, but rather regenerates the structure based on the preserved structural relationships. Furthermore, for example, the correspondence between table identifiers and column identifiers may be transformed according to substitution rules or mapping rules so that the original correspondence becomes difficult to deduce. Therefore, unlike information conversion technologies that process data content, the present invention is positioned as an information processing technology that preserves and regenerates structural topology information.

[0041] Figure 5 shows the concept of region-specific identification information and consistency determination according to the present invention.

[0042] During generation, at least one of the following is obtained from the environment in which the structure extraction means 11 and the mapping generation means 12 are executed: an OS identifier, a hardware identifier, location information, network identification information, etc. Figure 5 shows an example of this, including the OS identifier "Win11-xxxxx", hardware identifier "SN-12345", location information "34.5N,135.5E", and network identification information "192.168.1.0 / 24", but these are examples and are not limited to specific values. Based on this information, the region-specific identification information 17 at the time of generation is generated.

[0043] The mapping generation means 12 performs a mapping transformation using parameters based on the domain-specific identification information 17, and generates a non-business semantic representation that depends on the identification information. That is, the generated non-business semantic representation 32 not only maintains the structural topology but is also formed in a state associated with the execution environment at the time of generation. In this case, the generated domain-specific identification information 17 can be associated with the non-business semantic representation 32 or incorporated into the non-business semantic representation 32.

[0044] Subsequently, the non-business semantic expression 32 and the region-specific identification information 17 generated at that time are stored in the separate storage area by the storage means 13.

[0045] During reconstruction, the same type of identification information as at the time of generation is obtained from the current execution environment, and the current domain-specific identification information 21 is generated. The control means 16 compares the stored domain-specific identification information 17 at the time of generation with the current domain-specific identification information 21 obtained during reconstruction and performs a consistency determination.

[0046] The control means 16 permits the derivation and regeneration process by the reconstruction means 14 only if consistency is achieved, and prohibits the derivation of the structure from the non-business semantic expression 32 or the regeneration of the data structure if consistency is not achieved. In addition to the prohibition process, the control means 16 may also be configured to perform at least one of the following actions if consistency is not achieved: a rejection response, logging, a warning display, a re-authentication request, or an administrator notification.

[0047] The method for determining consistency is not limited and may include methods such as exact match determination, comparison of encoded values, hash matching, cryptographic verification, or partial match determination based on predetermined conditions.

[0048] Thus, the present invention is characterized by not simply saving the non-business semantic expression 32, but by associating it with the execution environment at the time of generation, thereby controlling the reconstruction process in an environment-dependent manner.

[0049] Figure 6 shows the concept of superimposing and analyzing identification patterns in the display output according to the present invention.

[0050] The display control means 19 outputs an identification pattern based on the region-specific identification information 17 (or coded information generated therefrom) superimposed on the display output based on the data structure derived and regenerated from the non-business semantic expression 32 by the reconstruction means 14. The display output screen 40 is output with the identification pattern superimposed on the business data display, and the identification pattern is embedded in a way that makes it difficult to see in the background or at the pixel level.

[0051] The identification pattern can be composed of high-frequency components, fine dots, periodic structures, or encoded pixel arrays. While it is not substantially visible under normal viewing conditions, it has a fragile characteristic (modification detection characteristic) in which parts of its structure are easily altered when image processing such as photography, resampling, or recompression is performed.

[0052] The acquired image 41 shows an example where the display output has been captured or recompressed, and a portion of the identification pattern is degraded. However, the identification pattern can be extracted by forensic analysis 42.

[0053] Forensic analysis 42 allows for the reconstruction of at least one of the following from the extracted identification pattern: region identification information, time information, session identification information, etc., based on a coded sequence that encodes region identification information or time information. This makes it possible to identify the source environment of the acquired image 41.

[0054] The format of the identification pattern, the embedding method, and the analysis method are not limited and may include spatial domain methods, frequency domain methods, watermarking techniques, digital watermarking techniques, steganography, or coded pattern methods.

[0055] Thus, the present invention is characterized by having, in addition to a structure restoration function, a function for detecting unauthorized acquisition and a function for identifying the source of output.

[0056] Figure 7A shows an example of the transformation of the graph structure in a mapping transformation according to the present invention. Figure 7A shows the original business structure 50, the topology structure 51 which is an abstraction of the business structure, and the reconstructed structure 52 which is generated based on the topology information.

[0057] The original business structure 50 shows an example of a data structure in a business system. In this example, there are three tables: "Orders," "Customers," and "Products," each with 5, 3, and 4 attribute columns, respectively. In addition, two connection relationships are formed by foreign key reference relationships.

[0058] The mapping generation means 12 performs a mapping transformation from the original business structure 50 to a non-business semantic representation 32 that retains structural relationships such as the number of nodes, the number of attributes for each node, the number of edges, and connection relationships, while removing or generalizing identifiers that have business meaning.

[0059] As shown in topology structure 51, each of the original tables is represented as anonymous nodes N1, N2, and N3, respectively, where the structural characteristics of the node are preserved by the number of attributes and the number of reference relationships, for example, N1 being "attr: 5, r: 3". Edges are maintained as connection relationships corresponding to the original foreign key reference relationships.

[0060] The reconstruction means 14 generates a restored structure 52 by assigning new identifiers, such as "Table_1", "Table_2", and "Table_3", based on the topology information held in the non-business semantic expression 32. The generated restored structure 52 is independent of the original business name, but is equivalent to the original structure in terms of the number of nodes, connection relationships, and attribute configuration.

[0061] The number of attributes, references, and names shown in the illustration are examples only, and the present invention is not limited to these.

[0062] Figure 7B is a diagram showing an example of a mapping transformation element according to the present invention, illustrating an example of the correspondence between elements retained and elements removed or generalized in the mapping transformation. Here, each element after mapping is not limited to individually replacing each of the original elements, but may be configured to be determined according to a combination of multiple structural features.

[0063] Identifiers with business meaning, such as table names and column names in the original business expression, are converted to non-business meaning expressions 32 after mapping. For example, "order" may be converted to an encoded identifier such as "x7f2a", and "customer_id" may be abstracted as "attr[1]".

[0064] On the other hand, structural elements such as the number of nodes (e.g., 3), edge relationships (e.g., 2), connection relationships, number of attributes, constraint types, and index structure are preserved. These elements form structural topology information as connection relationships between structural elements. For example, foreign key reference relationships may be expressed in the form of "N1.2 → N2.0", and constraint types may be preserved as encoded values.

[0065] The system can be configured to store only structural information, without saving actual row data, business details, transaction information, or other content information. Therefore, it is possible to retain only the structural information necessary for reconstruction while excluding customer names, addresses, transaction history, personal identification information, and other content information from storage.

[0066] Thus, the present invention is characterized by generating a reconstructible non-business semantic representation 32 by removing or generalizing business semantic information while retaining only structural relationships. It should be noted that the regeneration process in the present invention is not a process for restoring the stored business data content, but rather a process for deriving and regenerating the data structure based on the stored structural topology information.

[0067] Note that the identifier format, coding system, and storage method are examples only and are not limited to a specific format. In this embodiment, structural information includes information representing structural elements such as nodes, edges, connection relationships, number of attributes, constraint types, and their relationships. [Examples]

[0068] In this embodiment, we will describe an example in which the present invention is applied when a database file in a company's core business database (e.g., an order management system) is encrypted due to a ransomware attack, rendering the file contents unusable. First, when a business file is created or updated, structural information such as table definitions (order table, customer table, etc.), column configuration, reference relationships (foreign key constraints), constraint conditions, and index structure is extracted by the structure extraction means 11. Next, the mapping generation means 12 maps the structural information to a non-business semantic representation 32 in which business-meaning identifiers are removed or generalized, while maintaining the topology of the structural information. The generated non-business semantic representation 32 is stored by the storage means 13 in a storage area that is logically or physically separated from the original business file. If the contents of a database file are encrypted or lost due to a ransomware attack or the like, the reconstruction means 14 derives structural information from the non-business semantic representations 32 stored in the storage means 13, and regenerates the data structure on a new database based on the derived structural information. This reconstructs the database structure, including the table structure, reference relationships, and constraints. Thus, according to this embodiment, the data structure necessary for business continuity can be quickly regenerated without restoring the entire data content from backup data, as in the conventional method. As a result, for example, it becomes possible to prepare the database structure (so-called container) to which data from the backup medium will be input in a short time, contributing to the acceleration of system recovery work. Furthermore, according to this embodiment, even if the business data content itself is encrypted or corrupted, the data structure can be regenerated based on structural topology information without depending on the content. Therefore, it is possible to provide a structural recovery means that has high resistance to data content destruction by ransomware, etc. [Examples]

[0069] This embodiment describes an example of applying the present invention to the structural protection of business data in a cloud environment (e.g., a multi-tenant SaaS). For the data structure corresponding to the business data of each tenant, structural information such as table configuration, reference relationships, and constraints is extracted by the structure extraction means 11 and mapped to non-business semantic representations 32 by the mapping generation means 12. The generated non-business semantic representations 32 are stored by the storage means 13 in separate storage areas for each tenant. In this embodiment, tenant ID and cloud region information are used as domain-specific identifiers. During reconstruction, consistency checks are performed based on these domain-specific identifiers, and structural information is derived from the non-business semantic representation 32 only in matching environments, and the data structure is regenerated. This prevents the misuse of structural mapping information between tenants. Thus, according to this embodiment, data structures can be separated and stored on a tenant-by-tenant basis. For example, even if data for a specific tenant is lost due to a failure or other reason, it becomes possible to partially reconstruct only the data structure corresponding to that tenant. While conventional storage mechanisms in cloud environments (e.g., snapshots or backups) generally assume the restoration of data while retaining its content or semantic information, the present invention differs in that it extracts and stores information about the structural topology without saving the file content, and then regenerates the data structure based on that information. In other words, the present invention is characterized by its focus on retaining and regenerating structural information, rather than primarily on the transformation or transfer of data content. [Examples]

[0070] This embodiment describes an example of applying the present invention in a migration project from a legacy system such as AS / 400 to a state-of-the-art platform. First, structural information such as physical file definitions, logical file definitions, field configurations, and field reference relationships is extracted from the file structure of the legacy system by the structure extraction means 11. Next, the mapping generation means 12 maps the structural information to a non-business semantic representation 32, which is created by removing or generalizing business-related names or identifiers while maintaining the topology of the structural information. The generated non-business semantic representation 32 is saved by the storage means 13. If data loss or structural definition inconsistencies occur during the migration process, the reconstruction means 14 derives structural information from the saved non-business semantic representations 32 and regenerates the data structure on the destination platform based on the derived structural information. This makes it possible to reconstruct the data structure while preserving the file structure and reference relationships of the legacy system. [Examples]

[0071] This embodiment describes an example of applying the present invention to the structural protection of regulatory compliance data in the medical device manufacturing industry (e.g., clinical trial data management systems). Under a GxP (Good Practice) regulatory environment, ensuring data integrity and traceability is legally required. In this embodiment, the structure extraction means 11 extracts structural information such as the structure of the audit trail table, constraints of regulatory compliance fields, and reference relationships in the approval workflow. Next, the mapping generation means 12 maps the structural information to a non-business semantic representation 32, which removes or generalizes business-related names or identifiers while maintaining the topology of the structural information. The generated non-business semantic representation 32 is saved by the storage means 13. In the event of a system failure or data loss, the reconstruction means 14 derives structural information from the stored non-business semantic representations 32 and regenerates the data structure based on the derived structural information. Furthermore, an identification pattern can be superimposed on the display output based on the reconstructed data structure, thereby facilitating the tracking of data origin or the detection of misuse during audits. [Industrial applicability]

[0072] This invention is widely applicable to various business systems that handle structured data, such as databases, file servers, and cloud storage. In particular, it can be introduced as a configuration that complements conventional backup technologies as part of ransomware countermeasures, disaster recovery (DR), and business continuity planning (BCP).

[0073] Furthermore, the structural information mapping and reconstruction process according to the present invention can be used as a means to ensure the portability and reconstructibility of structural information in legacy system migration, database refactoring, cloud migration projects, and the like. In addition, the domain consistency control and the superposition of identification patterns on the display output can be used as technical elements for data governance, data loss prevention (DLP), and compliance. [Explanation of Symbols]

[0074] 10 Business file system, 11 Structure extraction means, 12 Mapping generation means, 13 Storage means, 14 Reconstruction means, 15 Consistency verification means, 16 Control means, 17 Region-specific identification information (at generation), 18 Restored file system, 19 Display control means, 20 Structure mapping reconstruction core, 21 Current region-specific identification information, 30 Original structure information, 31 Structure topology, 32 Non-business semantic representation, 40 Display output screen, 41 Acquired image, 42 Forensic analysis, 50 Original business structure, 51 Topological structure, 52 Restored structure

Claims

1. An information processing system that reconstructs the data structure after the loss of file contents, A structural extraction means that extracts structural information independent of the file's contents, A mapping generation means that, while maintaining the topology of the structural information, removes business-meaning identifiers or performs a mapping transformation to a generalized non-business-meaning representation, Storage means for storing the non-business semantic expression in a storage area that is logically or physically separated from the area where the file is stored, A reconstruction means for deriving structural information from the non-business semantic expression and regenerating the data structure based on the derived structural information, An information processing system characterized by comprising the following features.

2. The information processing system according to claim 1, characterized in that the topology retention retains at least one of the number of nodes, edge connection relationships, or constraint types.

3. The information processing system according to claim 1, characterized in that the mapping transformation includes abstraction, re-encoding, or transformation of correspondences between identifiers.

4. The information processing system according to claim 1, characterized in that the mapping transformation generates the non-business semantic representation by a mapping operation that includes at least one of the product term between the plurality of feature quantities, thresholding, or phase shift, based on the plurality of feature quantities constituting the structural information.

5. The information processing system according to claim 1, characterized in that the storage means does not store the contents of a file, but stores only the non-business semantic expression.

6. The information processing system according to claim 1, wherein the reconstruction means further comprises a consistency verification means for verifying the consistency of the structural constraints of the regenerated data structure.

7. The information processing system according to claim 1, further comprising a control means that generates the non-business semantic representation in association with the execution environment at the time of generation, based on the domain-specific identification information acquired at the time of generation, and executes the reconstruction means only when consistency with the domain-specific identification information is achieved.

8. The information processing system according to claim 7, characterized in that, if consistency with the aforementioned domain-specific identification information cannot be established, the derivation of the structural information or the regeneration of the data structure is prohibited.

9. The information processing system according to claim 7, characterized in that the region-specific identification information includes at least one of an OS identifier, a hardware identifier, location information, or network identification information.

10. The information processing system according to claim 7, further comprising a display control means that superimposes an identification pattern based on the area-specific identification information onto a display output based on the data structure regenerated by the reconstruction means.

11. The information processing system according to claim 10, characterized in that the identification pattern includes at least one of high-frequency components, fine dots, periodic structures, or encoded pixel arrays, and has a fragile property in which at least a portion of the identification pattern changes upon resampling or recompression.

12. The information processing system according to claim 10, characterized in that the identification pattern includes a code sequence that encodes region identification information or time information.

13. An information processing method executed by an information processing system that reconstructs a data structure after the loss of file contents, The structure extraction means of the information processing system includes a structure extraction step that extracts structure information independent of the file contents, The mapping generation means of the information processing system includes a mapping generation step of removing business-meaning identifiers or generalizing them into non-business-meaning representations while maintaining the topology of the structural information, The storage means of the information processing system includes a storage step of storing the non-business semantic expression in a storage area that is logically or physically separated from the area where the file is stored, The reconstruction means for the information processing system includes a reconstruction step of deriving structural information from the non-business semantic representation and regenerating the data structure based on the derived structural information, An information processing method characterized by including

14. A computer program for causing a computer to function as an information processing system according to any one of claims 1 to 12.

Images