Method and apparatus for constructing and applying an enterprise time-series knowledge graph driven by blockchain

Abstract

The present application relates to a method and device for constructing and applying a blockchain-driven enterprise temporal knowledge graph. The method comprises: deploying a seed temporal knowledge graph to the alliance blockchain network based on the underlying architecture to obtain a knowledge graph alliance chain; obtaining new enterprise information data contributed by enterprise users, and after uploading the data to the chain, increasing the points of enterprise users according to preset rules; the knowledge graph construction agency maintains and updates the temporal knowledge graph in the knowledge graph alliance chain; when receiving a service request from an enterprise user, the knowledge graph construction agency provides information services to the enterprise user based on the temporal knowledge graph in the knowledge graph alliance chain and a pre-built temporal reasoning model, and reduces the points of the enterprise user according to preset rules. The present invention constructs an open, trusted, and distributed knowledge graph construction architecture in which various enterprise entities actively participate, realizing a positive cycle of the platform.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method and apparatus for constructing and applying a blockchain-driven enterprise temporal knowledge graph. Background Art

[0002] As a domain-specific knowledge graph, an enterprise knowledge graph can encompass a wide range of information and data, including a company's operations, financial activities, legal affairs, and upstream and downstream supply chains. This resulting knowledge graph possesses significant economic value and social significance. It can provide business managers with reference information for operational decision-making, offer credit analysis to banks and other financial institutions, and provide government and research institutions with macro-level industry data and development trends. Building a comprehensive enterprise knowledge graph requires a vast amount of data, efficient algorithms, and continuous maintenance and updates. However, relying solely on a single organization to collect, build, and maintain such a vast amount of enterprise information presents challenges in data heterogeneity, data silos, and data security. These challenges include: 1. Data is multi-sourced and heterogeneous. Data is dispersed across many enterprises, making some enterprise data difficult to find online. Large enterprises are prone to data silos. 2. Most enterprises lack the internal drive to consistently share, update, and maintain knowledge graph data, making relying on a single or limited organization to collect data online insufficient. ③ After collecting data, independent organizations find it difficult to ensure that they do not add, delete, or modify it, rendering the resulting knowledge graph unreliable. ④ Storing knowledge graphs within a single organization also presents data security challenges. Hackers and other criminals can easily steal and modify data, leading to the leakage and unreliability of the underlying data in the knowledge graph. ⑤ Many application scenarios require maintaining the latest information and faster update speeds. Therefore, information in the time dimension is becoming increasingly important. How to develop time-series knowledge graph applications in the post-information era is also an important scientific issue that needs to be addressed urgently.

[0003] In summary, the existing technologies for constructing and applying enterprise knowledge graphs have problems with adaptability and poor security. Summary of the Invention

[0004] Based on this, it is necessary to provide a method, device, computer equipment and storage medium for constructing and applying a blockchain-driven enterprise temporal knowledge graph that can improve the adaptability and security of enterprise knowledge graph applications in response to the above technical problems.

[0005] A method for constructing and applying a blockchain-driven enterprise time-series knowledge graph, the method comprising:

[0006] Acquire enterprise information data, and construct a seed time-series knowledge graph based on the enterprise information data;

[0007] Based on the alliance blockchain network as the underlying architecture, the seed time-series knowledge graph is deployed to the alliance blockchain network to obtain a knowledge graph alliance chain; the network nodes of the knowledge graph alliance chain include enterprise users and knowledge graph construction institutions;

[0008] Obtain the new enterprise information data contributed by the enterprise user, confirm the data ownership based on smart contract technology, and after uploading the new enterprise information data to the knowledge graph alliance chain, increase the enterprise user's points value according to preset rules;

[0009] The knowledge graph construction organization updates the time-series knowledge graph in the knowledge graph alliance chain based on the knowledge graph alliance chain and the new enterprise information data on the chain;

[0010] When receiving a service request from an enterprise user, the knowledge graph construction organization will perform data authorization based on smart contract technology, provide information services to the enterprise user based on the temporal knowledge graph in the knowledge graph alliance chain and the pre-built temporal reasoning model, and reduce the enterprise user's points value according to preset rules.

[0011] In one embodiment, the further step includes: establishing ontology concept information of the enterprise application domain, and determining ontology classes based on the ontology concept information;

[0012] Acquire enterprise information data, obtain ontology-related data from the enterprise information data according to the ontology class, perform preprocessing, and obtain ontology knowledge; the ontology-related data includes time series information;

[0013] A seed temporal knowledge graph is constructed based on the ontology knowledge.

[0014] In one embodiment, it also includes: verifying the identity of the enterprise user on the chain through DID-based digital identity authentication technology.

[0015] In one embodiment, the further step includes: the knowledge graph construction organization issuing a graph update request;

[0016] Based on smart contract technology, new enterprise information data on the knowledge graph alliance chain is obtained;

[0017] Authenticate the ownership of the new enterprise information data and save the data ownership information on the knowledge graph alliance chain;

[0018] The time series knowledge graph in the knowledge graph alliance chain is updated.

[0019] In one embodiment, the information service further includes: information query service, decision analysis service, enterprise credit assessment service, enterprise risk assessment service and information question and answer service;

[0020] Correspondingly, the pre-built temporal reasoning model includes an information query model based on temporal reasoning, a decision analysis model based on temporal reasoning, an enterprise credit assessment model based on temporal reasoning, an enterprise risk assessment model based on temporal reasoning, and a question-answering model based on temporal reasoning.

[0021] In one embodiment, the enterprise credit assessment model based on temporal reasoning is a knowledge reasoning model based on ontology modeling, including: an enterprise credit assessment system construction module, a concept modeling module of an enterprise credit ontology, an inference rule definition module of an enterprise credit ontology, and an enterprise credit assessment module;

[0022] The enterprise credit assessment system construction module is used to construct an enterprise credit rating program;

[0023] The conceptual modeling module of the enterprise credit ontology is used to construct the enterprise credit ontology;

[0024] The inference rule definition module of the enterprise credit ontology is used to define inference rules based on ontology knowledge;

[0025] The enterprise credit evaluation module is used to obtain ontology knowledge from the time series knowledge graph according to the enterprise credit ontology, and output the enterprise credit evaluation result in combination with the enterprise credit rating scheme and the inference rules.

[0026] In one embodiment, the question-answering model based on temporal reasoning is a temporal reasoning model based on a subgraph, including an implicit expression parsing module, a relevant fact search module, and a subgraph logic reasoning module;

[0027] The implicit expression parsing module is used to obtain specific time constraint information of the problem based on the background knowledge in the temporal knowledge graph;

[0028] The related fact search module is used to search for candidate entities through temporal knowledge graph embedding according to the time constraint information to obtain an initial entity score;

[0029] The subgraph logic reasoning module is used to quantify the time constraint through temporal logic and obtain the final answer to the question according to the entity score.

[0030] In one embodiment, the further step further includes: obtaining the equity token of the enterprise user;

[0031] The knowledge graph construction organization constructs a data usage permission request transaction, signs it according to the predetermined conditional data format of the smart contract, sends it to the smart contract authorization interface, and uploads it to the knowledge graph alliance chain;

[0032] Automatically trigger the permission request smart contract, call the data owner's permission information, verify the legitimacy of the request, execute the relevant procedures according to the set code instructions, and give the response result. At the same time, the response result is recorded in the knowledge graph alliance chain, and the block information on the chain is broadcast in the network;

[0033] According to the response result of the smart contract, if the authorization is successful, the key used by the data owner to encrypt the data will be automatically sent to the account of the knowledge graph construction organization for encryption and use of the data.

[0034] A device for constructing and applying a blockchain-driven enterprise time-series knowledge graph, comprising:

[0035] A seed time series knowledge graph construction module is used to obtain enterprise information data and construct a seed time series knowledge graph based on the enterprise information data;

[0036] A knowledge graph alliance chain construction module is used to deploy the seed time-series knowledge graph to the alliance blockchain network based on the underlying architecture to obtain a knowledge graph alliance chain; the network nodes of the knowledge graph alliance chain include enterprise users and knowledge graph construction institutions;

[0037] The enterprise user data contribution module is used to obtain the new enterprise information data contributed by the enterprise user, confirm the data ownership based on smart contract technology, and increase the enterprise user's points value according to preset rules after uploading the new enterprise information data to the knowledge graph alliance chain;

[0038] A time-series knowledge graph updating module, configured for the knowledge graph construction mechanism to update the time-series knowledge graph in the knowledge graph alliance chain based on the knowledge graph alliance chain and new enterprise information data on the chain;

[0039] The service request response module is used to, when receiving a service request from an enterprise user, authorize data based on smart contract technology by the knowledge graph construction organization, provide information services to the enterprise user based on the temporal knowledge graph in the knowledge graph alliance chain and the pre-built temporal reasoning model, and reduce the enterprise user's point value according to preset rules.

[0040] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0041] Acquire enterprise information data, and construct a seed time-series knowledge graph based on the enterprise information data;

[0042] Based on the alliance blockchain network as the underlying architecture, the seed time-series knowledge graph is deployed to the alliance blockchain network to obtain a knowledge graph alliance chain; the network nodes of the knowledge graph alliance chain include enterprise users and knowledge graph construction institutions;

[0043] Obtain the new enterprise information data contributed by the enterprise user, confirm the data ownership based on smart contract technology, and after uploading the new enterprise information data to the knowledge graph alliance chain, increase the enterprise user's points value according to preset rules;

[0044] The knowledge graph construction organization updates the time-series knowledge graph in the knowledge graph alliance chain based on the knowledge graph alliance chain and the new enterprise information data on the chain;

[0045] When receiving a service request from an enterprise user, the knowledge graph construction organization will perform data authorization based on smart contract technology, provide information services to the enterprise user based on the temporal knowledge graph in the knowledge graph alliance chain and the pre-built temporal reasoning model, and reduce the enterprise user's points value according to preset rules.

[0046] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0047] Acquire enterprise information data, and construct a seed time-series knowledge graph based on the enterprise information data;

[0048] Based on the alliance blockchain network as the underlying architecture, the seed time-series knowledge graph is deployed to the alliance blockchain network to obtain a knowledge graph alliance chain; the network nodes of the knowledge graph alliance chain include enterprise users and knowledge graph construction institutions;

[0049] Obtain the new enterprise information data contributed by the enterprise user, confirm the data ownership based on smart contract technology, and after uploading the new enterprise information data to the knowledge graph alliance chain, increase the enterprise user's points value according to preset rules;

[0050] The knowledge graph construction organization updates the time-series knowledge graph in the knowledge graph alliance chain based on the knowledge graph alliance chain and the new enterprise information data on the chain;

[0051] When receiving a service request from an enterprise user, the knowledge graph construction organization will perform data authorization based on smart contract technology, provide information services to the enterprise user based on the temporal knowledge graph in the knowledge graph alliance chain and the pre-built temporal reasoning model, and reduce the enterprise user's points value according to preset rules.

[0052] The above-mentioned blockchain-driven enterprise time-series knowledge graph construction and application method, device, computer equipment and storage medium, uses the alliance blockchain network as the underlying architecture to deploy the seed time-series knowledge graph to the alliance blockchain network to obtain the knowledge graph alliance chain; obtains new enterprise information data contributed by enterprise users, and after uploading the data to the knowledge graph alliance chain based on smart contract technology, increases the enterprise user's points according to preset rules; the knowledge graph construction agency maintains and updates the time-series knowledge graph in the knowledge graph alliance chain; when receiving a service request from an enterprise user, the knowledge graph construction agency performs data authorization based on smart contract technology, provides information services to the enterprise user based on the time-series knowledge graph in the knowledge graph alliance chain and the pre-built time-series reasoning model, and reduces the enterprise user's points according to preset rules. The present invention allows enterprises, banks and other institutions to join as nodes to jointly build and share data in the enterprise knowledge graph and achieve a win-win situation in application services. The knowledge graph uses a temporal knowledge graph, adding the characteristics of the time dimension to the traditional knowledge graph. On this basis, a variety of temporal reasoning models are constructed to provide support for decision analysis, risk assessment, question-answering systems and other fields, and provide external information services. This invention constructs an open, distributed, and trustworthy knowledge graph construction architecture with the active participation of various enterprise entities, realizing the positive cycle of the platform, promoting the updating and iteration of the temporal knowledge graph, and continuously enhancing the platform's knowledge service capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A flowchart of a method for constructing and applying a blockchain-driven enterprise time-series knowledge graph in one embodiment;

[0054] Figure 2 A technical architecture diagram for the construction and application of a blockchain-driven enterprise time-series knowledge graph in one embodiment;

[0055] Figure 3 This is a diagram of the overall architecture of a blockchain-based enterprise trusted knowledge graph in a method for constructing and applying a blockchain-driven enterprise time-series knowledge graph in one embodiment;

[0056] Figure 4 A structural block diagram of an apparatus for constructing and applying a blockchain-driven enterprise time-series knowledge graph in one embodiment;

[0057] Figure 5FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0059] In one embodiment, Figure 1 As shown, a method for constructing and applying a blockchain-driven enterprise time series knowledge graph is provided, including the following steps:

[0060] Step 102: Obtain enterprise information data and construct a seed time series knowledge graph based on the enterprise information data.

[0061] First, we need to complete the construction of the initial seed time-series knowledge graph. The construction of the seed time-series knowledge graph mainly includes three steps: first, establishing the concept layer of the enterprise domain ontology, second, collecting relevant data based on the ontology class and performing data preprocessing, and third, establishing a seed knowledge graph for the ontology knowledge. The data used is the data of more than 4,000 A-share listed companies for the construction of the seed enterprise knowledge graph. Its data advantage lies in the fact that it covers almost all industries, has a wide range of business coverage, and has rich and reliable data sources. Data features include basic information of the enterprise such as company name, legal person, operating status, establishment time (time-series features), registered capital, paid-in capital, business term (time-series features), company type, industry, and staff size.

[0062] During the construction of the seed enterprise knowledge graph, iterations involve data expansion and graph mapping on the blockchain, enabling the construction of a trustworthy, large-scale graph in a small sample environment. This means that blockchain-based iteration of the seed knowledge graph is a process of continuous association, integration, and updating of trusted knowledge. Building a seed knowledge graph is crucial for subsequent tasks. On the one hand, the seed graph contains a large number of standardized entities, providing a reference for subsequent entity alignment. On the other hand, using the seed knowledge graph to align unlabeled data is essentially an application of a remote supervision algorithm, effectively reducing data annotation costs.

[0063] The knowledge graph of the present invention adopts a time-series knowledge graph, which adds the characteristics of the time dimension to the traditional knowledge graph and enhances the timeliness of the enterprise knowledge graph.

[0064] Step 104: Based on the alliance blockchain network as the underlying architecture, the seed time series knowledge graph is deployed to the alliance blockchain network to obtain a knowledge graph alliance chain.

[0065] The network nodes of the knowledge graph alliance chain include enterprise users and knowledge graph construction institutions. Knowledge graph construction institutions can be financial institutions such as banks, or governments, scientific research institutions, etc.

[0066] A consortium chain refers to a blockchain network jointly maintained by multiple institutions. Each institution joins or exits the network after authorization. The permissions of each institution are completely equal, and the reading, writing, and accounting rules of on-chain information are determined according to the consortium consensus.

[0067] This invention uses the alliance blockchain network as the underlying architecture, deploys the enterprise knowledge graph on the blockchain, and allows enterprises, banks and other institutions to join as nodes to jointly build and share data in the enterprise knowledge graph and achieve win-win application services.

[0068] All nodes on the chain are responsible for the operation of the consensus mechanism to ensure the validity of the client node (enterprise user node), the integrity and consistency of the data of each node, and to prevent attacks from malicious Byzantine nodes.

[0069] Step 106: obtain the new enterprise information data contributed by the enterprise user, confirm the data ownership based on smart contract technology, and upload the new enterprise information data to the knowledge graph alliance chain, and then increase the enterprise user's points value according to the preset rules.

[0070] As blockchain nodes and clients of the Knowledge Graph Alliance, enterprise users are responsible for maintaining the operation of the Knowledge Graph blockchain system and providing company-related data, which serves as a key component of knowledge graph construction and updates. Furthermore, by participating in maintenance and construction, enterprise users can earn equity tokens in the system, or points in this embodiment, which can be used to request and implement value-added services. The enterprise user module primarily interacts with the Knowledge Graph Alliance chain through smart contract technology, including data contribution and service requests.

[0071] The data on-chain process utilizes data ownership confirmation technology. During the confirmation process, blockchain is used to preserve data ownership, ensuring the decentralization, traceability, immutability, and permanent preservation of data ownership records. Symmetric and asymmetric encryption algorithms are used to encrypt and sign the original data, ensuring the confidentiality of the original data while confirming data ownership.

[0072] In order to increase the participation of enterprise users in updating and maintaining relevant data on the chain, the present invention will reward enterprise users with a certain amount of points tokens each time the data information they upload passes the consensus mechanism. The tokens cannot be used to exchange for legal currency, but can be used to exchange for information query services, visual analysis services, recommendation systems, corporate credit assessment services, etc. built based on the knowledge graph on the chain. This is the inherent driving force for enterprises to join and share information. Relevant institutions can also develop similar information services on this chain. After the developer publishes the service, when other user nodes use the service, they also need to transfer the token to the developer node. All token generation, circulation, conversion, and conversion processes are automatically executed through the on-chain smart contract to ensure the fairness and credibility of the process.

[0073] In step 108, the knowledge graph construction organization updates the time-series knowledge graph in the knowledge graph alliance chain based on the knowledge graph alliance chain and the new enterprise information data on the chain.

[0074] As a blockchain node and graph construction and service platform within the Knowledge Graph Alliance, the graph construction organization is responsible for maintaining the operation of the alliance chain. It also needs to build and regularly update the enterprise knowledge graph based on the data uploaded by each client. It also provides high-quality value-added services in a timely manner after obtaining the user's equity token.

[0075] Step 110: When a service request is received from an enterprise user, the knowledge graph construction organization performs data authorization based on smart contract technology, provides information services to the enterprise user based on the temporal knowledge graph in the knowledge graph alliance chain and the pre-built temporal reasoning model, and reduces the enterprise user's points value according to preset rules.

[0076] The reduced points value can be transferred to the account of the knowledge graph construction organization node.

[0077] Compared with the construction of corporate credit knowledge graphs by centralized service platforms, the construction of open credit knowledge graphs with the active participation of various corporate groups better reflects fairness and trust.

[0078] like Figure 2 This is a technical architecture diagram of this embodiment, including the blockchain network bottom layer, natural language processing and knowledge modeling layer, algorithm layer, blockchain interaction layer and application layer. Figure 3This is the overall architecture of a blockchain-based enterprise trusted knowledge graph. It consists of three components: the knowledge graph consortium chain, enterprise users, and the graph construction organization. The knowledge graph consortium chain is based on a consortium blockchain network. Network nodes include both willing enterprise users and knowledge graph construction organizations. All nodes on the chain are responsible for running the consensus mechanism to ensure the validity of client node identities, data integrity and consistency across nodes, and prevent attacks from malicious Byzantine nodes. This module primarily includes blockchain-based data sharing technology. Enterprise users are the blockchain nodes of the knowledge graph consortium and the clients of the system. This module primarily interacts with the knowledge graph consortium chain through smart contracts, including data contribution and service requests. The graph construction organization comprises the blockchain nodes of the knowledge graph consortium and the graph construction and service platform. This module primarily involves intelligent reasoning technology based on time-series knowledge graphs.

[0079] In the above-mentioned blockchain-driven enterprise time-series knowledge graph construction and application method, the seed time-series knowledge graph is deployed to the alliance blockchain network with the alliance blockchain network as the underlying architecture to obtain a knowledge graph alliance chain; new enterprise information data contributed by enterprise users is obtained, and after the data is uploaded to the knowledge graph alliance chain based on smart contract technology, the enterprise user's point value is increased according to preset rules; the knowledge graph construction organization maintains and updates the time-series knowledge graph in the knowledge graph alliance chain; when a service request from an enterprise user is received, the knowledge graph construction organization performs data authorization based on smart contract technology, and provides information services to the enterprise user based on the time-series knowledge graph in the knowledge graph alliance chain and the pre-built time-series reasoning model, and reduces the enterprise user's point value according to preset rules.

[0080] The technical effect achieved by the present invention is: allowing enterprises, banks and other institutions to join as nodes, to jointly build and share data in the enterprise knowledge graph, and to achieve a win-win situation in application services. By adopting a time-series knowledge graph, the characteristics of the time dimension are added to the traditional knowledge graph, and a variety of time-series reasoning models are constructed on this basis to provide support for decision analysis, risk assessment, question-answering systems and other fields, and to provide information services to the outside world. The present invention constructs an open, distributed, trustworthy knowledge graph construction architecture in which various corporate entities actively participate. It has the technical characteristics of authentic identity identification, reliable credit sources, fair incentives and rewards, transparent collaborative processes, and trustworthy service creation. It innovatively solves the problem of enterprise data collection, helps enterprises to conduct situational awareness, risk analysis, auxiliary decision-making, etc., and thus realizes cost reduction and efficiency improvement for enterprises.

[0081] In one of the embodiments, it also includes: verifying the identity of corporate users on the chain through DID-based digital identity authentication technology.

[0082] Traditional identity authentication methods rely on trusted centralized institutions, posing risks such as vulnerability to tampering, malicious attacks, and single points of failure. Taking into account the requirements for decentralization, interoperability, privacy protection, and on-chain riveting of digital identities in blockchain-based knowledge graph consortium chains, this embodiment uses blockchain-based DID (Decentralized IDentity) distributed digital identification technology to construct enterprise user IDs, ensuring trusted mapping between enterprise entities and on-chain objects, as well as secure access authorization and data sharing between entity objects in a decentralized environment.

[0083] The specific process is as follows:

[0084] Step 1: Generate keys and DID identifiers. The enterprise first generates a public-private key pair locally through an asymmetric encryption function.<PubKey,PriKey> , where the private key PriKey will be stored locally on the enterprise user and will not be made public; the DID identifier can be obtained by applying to the W3C website maintained by the graph construction organization.

[0085] Step 2: Submit information. The enterprise assembles an identity authentication request, including its public key, DID, and identity verification materials, and sends it to the DID certification authority. The authority reviews and verifies the DID (the digital identity identifier applied for by the enterprise) and its associated detailed identity information, including the set of encryption materials (such as the public key) and the set of authentication materials. It then generates a DID document that details the DID and other related information about the enterprise.

[0086] Step 3: On-chain storage and delivery. After approval by the DID certification authority, the DID document is signed using its own private key DID identifier and sent to the blockchain for consensus storage. After successful on-chain storage, the DID document is sent to the enterprise.

[0087] Step 4: Apply for VC certification. The enterprise sends the DID document to an authoritative agency, such as the Administration for Industry and Commerce. After the Administration compares and verifies the DID digital identity stored on the chain, if it is correct, it generates VC certification materials, stores them on the chain successfully, and sends them to the enterprise.

[0088] Step 5: The enterprise sends the DID digital identity materials and VC certification materials to the graph construction agency. After the graph construction agency verifies that the stored information on the blockchain is correct, it opens the interface of the knowledge graph alliance chain to the enterprise, and the enterprise uses its own digital identity to enter the alliance chain.

[0089] In one of the embodiments, it also includes: the knowledge graph construction organization issues a graph update request; obtains new enterprise information data on the knowledge graph alliance chain based on smart contract technology; authenticates the data ownership of the new enterprise information data, and saves the data ownership information on the knowledge graph alliance chain; and updates the time-series knowledge graph in the knowledge graph alliance chain.

[0090] In one embodiment, it also includes: information services include information query services, decision analysis services, enterprise credit assessment services, enterprise risk assessment services and information question and answer services; correspondingly, the pre-built temporal reasoning model includes an information query model based on temporal reasoning, a decision analysis model based on temporal reasoning, an enterprise credit assessment model based on temporal reasoning, an enterprise risk assessment model based on temporal reasoning and a question and answer model based on temporal reasoning.

[0091] In one embodiment, it also includes: the enterprise credit assessment model based on temporal reasoning is a knowledge reasoning model based on ontology modeling, including: an enterprise credit assessment system construction module, a conceptual modeling module of an enterprise credit ontology, an inference rule definition module of an enterprise credit ontology and an enterprise credit assessment module; the enterprise credit assessment system construction module is used to construct an enterprise credit rating scheme; the conceptual modeling module of the enterprise credit ontology is used to construct an enterprise credit ontology; the inference rule definition module of the enterprise credit ontology is used to define inference rules based on ontology knowledge; the enterprise credit assessment module is used to obtain ontology knowledge from a temporal knowledge graph according to the enterprise credit ontology, and output the enterprise credit assessment results in combination with the enterprise credit rating scheme and inference rules.

[0092] Knowledge reasoning is the process of inferring unknown knowledge based on existing knowledge. This is done by starting from known knowledge, using the knowledge already acquired, and obtaining new facts contained therein, or generalizing from a large amount of existing knowledge, from individual knowledge to general knowledge.

[0093] Credit reporting is the legal collection, organization, storage, and processing of credit information on individuals, legal entities, and other organizations, along with the provision of credit reports, credit assessments, and credit consulting services to help clients assess and control credit risk and conduct credit management. The fundamental functions of corporate credit reporting include preventing credit risk, expanding credit transactions, improving economic efficiency, and promoting the development of a social credit system.

[0094] As a complete system, corporate credit rating encompasses three aspects: credit rating indicators, credit rating grading standards, and credit rating methodologies. Credit rating indicators and credit rating methodologies are the most core, closely linked, and influential components of the credit rating system. This embodiment primarily builds a corporate credit ontology model by referencing the indicators of existing corporate credit assessment systems, thereby enabling intelligent reasoning about corporate credit ratings.

[0095] With reference to the grading standards of the National Enterprise Credit Assessment Network, the credit rating can be divided as shown in the following table (this rating is only for reference and experimental basis, and in actual application, more detailed divisions can be made according to needs):

[0096] Table 1 Enterprise credit rating classification

[0097]

[0098] To implement corporate credit rating in the ontology model, a series of detailed indicators and calculation logic need to be referenced to obtain the final credit rating results. This paper screens, optimizes, and abstracts existing indicators to construct a corporate credit rating scheme that is suitable for ontology modeling, as shown in Table 2 below:

[0099] Table 2 Enterprise Credit Rating Scheme

[0100]

[0101]

[0102] Note: Credit record risk (high, medium, low), enterprise development status (excellent, good, poor), legal litigation risk (high, medium, low), enterprise soft power (strong, medium, weak); Represents the logical symbol "not", and Any represents any value within the optional definition domain.

[0103] The steps to build an enterprise credit ontology are as follows:

[0104] Stage 1: Determine the ontology domain as corporate credit investigation

[0105] Based on the company's financial information, legal information, business operations, intellectual property rights, industry status and other data, we build a corporate credit ontology to solve the problem of corporate credit assessment.

[0106] Stage 2: Find and reuse ontologies from open resources

[0107] I found two reference resources on the OpenKG open source platform: the enterprise knowledge graph and the innovation investment field knowledge graph. I referred to some of the classes and data defined in them, such as enterprise, financial, (bid) contract, patent, credit, tax and other data.

[0108] Stage 3: Some important concepts of corporate credit ontology

[0109] The entity categories that must be included in the corporate credit ontology include the enterprise itself, corporate business events, enterprise-related personnel and organizations, qualifications and intellectual property rights owned by the enterprise, etc.; the enterprise's data attributes include basic corporate information, business information, financial information, etc.; the relationship attributes in the ontology include the relationship between the enterprise and products, people, other organizations, patents, bids, legal documents, announcements and other things.

[0110] Stage 4: Define classes and relationships between classes

[0111] Use protégé to complete the construction of corporate credit ontology Classes.

[0112] Stage 5: Define external and internal attributes

[0113] Use protégé to complete the construction of Object properties and Data properties of the corporate credit ontology.

[0114] Stage 6: Define attribute constraints

[0115] In Protégé, Domains, Ranges, and Characteristics are defined based on the head and tail entity categories of each relationship attribute. Similarly, Domains and Ranges are defined based on the entity objects described by each data attribute.

[0116] Stage7: Creating and importing instances

[0117] Based on the completed ontology concepts and the collected multi-source data, the data schema is constructed and finally imported into the protégé tool using the Cellfie tool.

[0118] An inference rule can be a sentence that treats any statement as an "if...then..." sentence. Any type of rule contains at least one premise and one conclusion. In description logic, an inference rule is a scheme for constructing valid inferences.

[0119] The reasoning rules in the corporate credit ontology are divided into two stages. The first stage mainly uses 12 rules to describe the classification basis of each sub-category within the four major categories: credit record risk (high, medium, and low), corporate development status (excellent, good, and poor), legal litigation risk (high, medium, and low), and corporate soft power (strong, medium, and weak). Among them, the level of credit record risk is mainly determined by the number of credit defaults and the total amount of credit defaults; corporate development status is mainly determined by the compound growth rate of net profit; legal litigation risk is mainly determined by the number of defendant lawsuits and the total amount of defendant lawsuits; and corporate soft power is determined by the number of invention patents, the number of corporate qualifications, and the number of government awards received.

[0120] Taking the credit record risk reasoning as an example, we elaborate on the following rules: ① If a company's credit default count is ≤ 1 and the default amount is ≤ 1 million, it is classified as a low-risk company; ② If a company's credit default count is 1 and 1 million < the default amount ≤ 10 million, it is classified as a medium-risk company; ③ If a company's credit default count is > 1 or the default amount is > 10 million, it is classified as a high-risk company. This can be expressed in SWRL as follows:

[0121] Table 3 Credit record risk SWRL inference rules

[0122]

[0123] The second stage, building on the foundation of the first stage, expresses inference rules for the enterprise credit rating scheme in Table 4. This scheme encompasses 81 scenarios and 12 rules. Enterprises are categorized into five grades: AAA, AA, A, B, and C. Taking B as an example, the rule for this scheme describes the following: If an enterprise's credit risk category is low, its development is non-excellent, its legal risk is non-low, and its soft strength is non-strong, then its credit rating is B. Alternatively, if an enterprise's credit risk category is medium and it does not exist in the intersection of poor development, high legal risk, or weak soft strength, then its rating is also B.

[0124] Table 4 SWRL inference rules for Class B credit reporting companies

[0125]

[0126] After completing the ontology concept modeling, instance importing, and rule writing in protégé, the HermiT inference engine is executed to complete the credit rating reasoning of the enterprise instance and the consistency test of the ontology model. By statistically analyzing the various situations in Table 3, the enterprise credit rating scheme table, it can also be proved that the ontology model has passed the completeness test.

[0127] Taking the three corporate discovery results "Shensi Electronics", "China Power Construction", and "China Merchants Bank" as examples, we queried them in Protégé and obtained the inference results shown in the following table. Based on the actual situation of the companies, their credit ratings were analyzed and found to be basically consistent with the description of the results inferred using the ontology, verifying the effectiveness of the corporate credit ontology model.

[0128] Table 5 Verification of enterprise credit ontology reasoning results

[0129]

[0130] In one of the embodiments, it also includes: the question-answering model based on temporal reasoning is a subgraph-based temporal reasoning model, including: an implicit expression parsing module, a relevant fact search module and a subgraph logical reasoning module; the implicit expression parsing module is used to obtain the specific time constraint information of the question based on the background knowledge in the temporal knowledge graph; the relevant fact search module is used to embed the search for candidate entities through the temporal knowledge graph according to the time constraint information to obtain an initial entity score; the subgraph logical reasoning module is used to quantify the time constraints through temporal logic and obtain the final answer to the question based on the entity score.

[0131] This embodiment proposes a subgraph-based model for answering complex questions on temporal knowledge graphs (TKGs), called subgraph temporal reasoning (SubGTR), which consists of three main modules: implicit knowledge extraction, relevant fact search, and subgraph logical reasoning.

[0132] In the implicit expression parsing module, the temporal information hidden in the question is extracted and parsed using knowledge from the temporal knowledge graph. Regular expressions are used to replace the natural language question with a simplified version. Time constraints T1 and T2 are then added to the question for subsequent reasoning and querying. In the related fact search module, the natural language question is first converted into a vector representation that combines the question semantics, the temporal knowledge graph semantics, and the temporal semantics. Then, the TKG scoring function and a time-sensitive function are used to obtain candidate entities and their semantic scores. In the subgraph temporal reasoning module, a subgraph is first constructed for each question, which is then pruned using temporal constraints. Finally, a temporal activation function is used to quantitatively score each entity in the subgraph, and the final answer is obtained by fusing the scores.

[0133] In one of the embodiments, it also includes: obtaining the equity token of the enterprise user; the knowledge graph construction organization constructs a data usage permission request transaction, signs and sends it to the smart contract authorization interface in accordance with the predetermined conditional data format of the smart contract, and uploads it to the knowledge graph alliance chain; automatically triggers the permission request smart contract, calls the data owner's permission information, verifies the legitimacy of the request, executes the relevant procedures according to the set code instructions, and gives a response result, and at the same time records the response result in the knowledge graph alliance chain, and broadcasts the block information on the chain in the network; according to the response result of the smart contract, if the authorization is successful, the data owner's key used to encrypt the data is automatically sent to the account of the knowledge graph construction organization for data encryption and use.

[0134] After an enterprise uploads data or information to a blockchain network, authorization for its use is required to prevent misuse. This invention employs automated authorization technology based on smart contracts, further improving the efficiency, security, and fairness of data engineering and usage.

[0135] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0136] In one embodiment, Figure 4 As shown, a device for constructing and applying a blockchain-driven enterprise time-series knowledge graph is provided, including: a seed time-series knowledge graph construction module 402, a knowledge graph alliance chain construction module 404, an enterprise user data contribution module 406, a time-series knowledge graph update module 408, and a service request response module 410, wherein:

[0137] A seed time series knowledge graph construction module 402 is used to obtain enterprise information data and construct a seed time series knowledge graph based on the enterprise information data;

[0138] The knowledge graph alliance chain construction module 404 is used to deploy the seed time series knowledge graph to the alliance blockchain network based on the underlying architecture to obtain a knowledge graph alliance chain; the network nodes of the knowledge graph alliance chain include enterprise users and knowledge graph construction institutions;

[0139] The enterprise user data contribution module 406 is used to obtain new enterprise information data contributed by enterprise users, confirm data ownership based on smart contract technology, and increase the enterprise user's points according to preset rules after uploading the new enterprise information data to the knowledge graph alliance chain;

[0140] The time series knowledge graph update module 408 is used for the knowledge graph construction organization to update the time series knowledge graph in the knowledge graph alliance chain based on the knowledge graph alliance chain and the new enterprise information data on the chain;

[0141] The service request response module 410 is used to authorize data based on smart contract technology when receiving a service request from an enterprise user. It provides information services to the enterprise user based on the temporal knowledge graph in the knowledge graph alliance chain and the pre-built temporal reasoning model, and reduces the enterprise user's point value according to preset rules.

[0142] The seed temporal knowledge graph construction module 402 is also used to establish ontology concept information of the enterprise application field, determine the ontology class based on the ontology concept information; obtain enterprise information data, obtain ontology-related data from the enterprise information data according to the ontology class for preprocessing, and obtain ontology knowledge; the ontology-related data includes temporal information; and construct a seed temporal knowledge graph based on the ontology knowledge.

[0143] The enterprise user data contribution module 406 is also used to verify the identity of enterprise users on the chain through DID-based digital identity authentication technology.

[0144] The time-series knowledge graph update module 408 is also used by the knowledge graph construction organization to issue a graph update request; obtain new enterprise information data on the knowledge graph alliance chain based on smart contract technology; authenticate the data ownership of the new enterprise information data, and save the data ownership information on the knowledge graph alliance chain; and update the time-series knowledge graph in the knowledge graph alliance chain.

[0145] The service request response module 410 is also used to obtain the equity token of the enterprise user; the knowledge graph construction organization constructs the data usage permission request transaction, signs and sends it to the smart contract authorization interface in accordance with the predetermined conditional data format of the smart contract, and uploads it to the knowledge graph alliance chain; automatically triggers the permission request smart contract, calls the data owner's permission information, verifies the legitimacy of the request, executes the relevant procedures according to the set code instructions, and gives the response result, and at the same time records the response result in the knowledge graph alliance chain, and broadcasts the block information on the chain in the network; according to the response result of the smart contract, if the authorization is successful, the data owner's key used to encrypt the data is automatically sent to the knowledge graph construction organization's account for data encryption and use.

[0146] Regarding the specific definition of the device for constructing and applying a blockchain-driven enterprise time series knowledge graph, please refer to the definition of the method for constructing and applying a blockchain-driven enterprise time series knowledge graph above, which will not be repeated here. Each module in the above-mentioned device for constructing and applying a blockchain-driven enterprise time series knowledge graph can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0147] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, memory, network interface, display screen, and input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements a method for constructing and applying a blockchain-driven enterprise time-series knowledge graph. The display screen of the computer device can be a liquid crystal display or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or it can be a button, trackball, or touchpad provided on the computer device housing, or it can be an external keyboard, touchpad, or mouse.

[0148] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0149] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiment when executing the computer program.

[0150] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0151] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0152] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0153] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A blockchain-driven enterprise time series knowledge graph construction and application method, characterized by: The method comprises: Acquire enterprise information data, and construct a seed time-series knowledge graph based on the enterprise information data; Based on the alliance blockchain network as the underlying architecture, the seed time-series knowledge graph is deployed to the alliance blockchain network to obtain a knowledge graph alliance chain; the network nodes of the knowledge graph alliance chain include enterprise users and knowledge graph construction institutions; Obtain the new enterprise information data contributed by the enterprise user, confirm the data ownership based on smart contract technology, and after uploading the new enterprise information data to the knowledge graph alliance chain, increase the enterprise user's points value according to preset rules; The knowledge graph construction organization updates the time-series knowledge graph in the knowledge graph alliance chain based on the knowledge graph alliance chain and the new enterprise information data on the chain; When receiving a service request from an enterprise user, the knowledge graph construction organization performs data authorization based on smart contract technology, provides information services to the enterprise user based on the time series knowledge graph in the knowledge graph alliance chain and the pre-built time series reasoning model, and reduces the enterprise user's points according to preset rules; Acquire enterprise information data and construct a seed time-series knowledge graph based on the enterprise information data, including: Establishing ontology concept information of the enterprise application domain, and determining ontology classes based on the ontology concept information; Acquire enterprise information data, obtain ontology-related data from the enterprise information data according to the ontology class, perform preprocessing, and obtain ontology knowledge; the ontology-related data includes time series information; Constructing a seed temporal knowledge graph based on the ontology knowledge; Before obtaining the new enterprise information data contributed by the enterprise user, confirming the data ownership based on smart contract technology, and uploading the new enterprise information data to the knowledge graph alliance chain, the method further includes: Verify the identity of the enterprise user on the chain through DID-based digital identity authentication technology; The knowledge graph construction organization updates the time series knowledge graph in the knowledge graph alliance chain based on the knowledge graph alliance chain and the new enterprise information data on the chain, including: The knowledge graph construction organization issues a graph update request; Based on smart contract technology, new enterprise information data on the knowledge graph alliance chain is obtained; Authenticate the ownership of the new enterprise information data and save the data ownership information on the knowledge graph alliance chain; The time series knowledge graph in the knowledge graph alliance chain is updated.

2. The method according to claim 1, characterized in that The information services include information query services, decision analysis services, enterprise credit assessment services, enterprise risk assessment services and information question and answer services; Correspondingly, the pre-built temporal reasoning model includes an information query model based on temporal reasoning, a decision analysis model based on temporal reasoning, an enterprise credit assessment model based on temporal reasoning, an enterprise risk assessment model based on temporal reasoning, and a question-answering model based on temporal reasoning.

3. The method according to claim 2, characterized in that The enterprise credit assessment model based on temporal reasoning is a knowledge reasoning model based on ontology modeling, including: an enterprise credit assessment system construction module, a concept modeling module of the enterprise credit ontology, an inference rule definition module of the enterprise credit ontology, and an enterprise credit assessment module; The enterprise credit assessment system construction module is used to construct an enterprise credit rating program; The conceptual modeling module of the enterprise credit ontology is used to construct the enterprise credit ontology; The inference rule definition module of the enterprise credit ontology is used to define inference rules based on ontology knowledge; The enterprise credit evaluation module is used to obtain ontology knowledge from the time series knowledge graph according to the enterprise credit ontology, and output the enterprise credit evaluation result in combination with the enterprise credit rating scheme and the inference rules.

4. The method according to claim 2, characterized in that The temporal reasoning-based question-answering model is a subgraph-based temporal reasoning model, comprising: an implicit expression parsing module, a relevant fact search module, and a subgraph logic reasoning module; The implicit expression parsing module is used to obtain specific time constraint information of the problem based on the background knowledge in the temporal knowledge graph; The related fact search module is used to search for candidate entities through temporal knowledge graph embedding according to the time constraint information to obtain an initial entity score; The subgraph logic reasoning module is used to quantify the time constraint through temporal logic and obtain the final answer to the question according to the entity score.

5. The method according to claim 1, wherein When receiving a service request from an enterprise user, the knowledge graph construction organization performs data authorization based on smart contract technology, which also includes: Obtain the equity token of the enterprise user; The knowledge graph construction organization constructs a data usage permission request transaction, signs it according to the predetermined conditional data format of the smart contract, sends it to the smart contract authorization interface, and uploads it to the knowledge graph alliance chain; Automatically trigger the permission request smart contract, call the data owner's permission information, verify the legitimacy of the request, execute the relevant procedures according to the set code instructions, and give the response result. At the same time, the response result is recorded in the knowledge graph alliance chain, and the block information on the chain is broadcast in the network; According to the response result of the smart contract, if the authorization is successful, the key used by the data owner to encrypt the data will be automatically sent to the account of the knowledge graph construction organization for encryption and use of the data.

6. A device for constructing and applying a blockchain-driven enterprise time-series knowledge graph, characterized in that: The device comprises: A seed temporal knowledge graph construction module is used to obtain enterprise information data and construct a seed temporal knowledge graph based on the enterprise information data, including: establishing ontology concept information of the enterprise application field and determining ontology classes based on the ontology concept information; obtaining enterprise information data, obtaining ontology-related data from the enterprise information data according to the ontology classes, performing preprocessing to obtain ontology knowledge; the ontology-related data includes temporal information; and constructing a seed temporal knowledge graph based on the ontology knowledge; A knowledge graph alliance chain construction module is used to deploy the seed time-series knowledge graph to the alliance blockchain network based on the underlying architecture to obtain a knowledge graph alliance chain; the network nodes of the knowledge graph alliance chain include enterprise users and knowledge graph construction institutions; The enterprise user data contribution module is used to obtain the new enterprise information data contributed by the enterprise user, confirm the data ownership based on smart contract technology, and increase the enterprise user's points value according to preset rules after the new enterprise information data is uploaded to the knowledge graph alliance chain; before obtaining the new enterprise information data contributed by the enterprise user, confirming the data ownership based on smart contract technology, and uploading the new enterprise information data to the knowledge graph alliance chain, it also includes: Verify the identity of the enterprise user on the chain through DID-based digital identity authentication technology; The time series knowledge graph update module is used by the knowledge graph construction mechanism to update the time series knowledge graph in the knowledge graph alliance chain based on the knowledge graph alliance chain and the new enterprise information data on the chain, including: The knowledge graph construction organization issues a graph update request; Based on smart contract technology, new enterprise information data on the knowledge graph alliance chain is obtained; Authenticate the ownership of the new enterprise information data and save the data ownership information on the knowledge graph alliance chain; Updating the time series knowledge graph in the knowledge graph alliance chain; The service request response module is used to, when receiving a service request from an enterprise user, authorize data based on smart contract technology by the knowledge graph construction organization, provide information services to the enterprise user based on the temporal knowledge graph in the knowledge graph alliance chain and the pre-built temporal reasoning model, and reduce the enterprise user's point value according to preset rules.

7. The device according to claim 6, characterized in that The seed temporal knowledge graph construction module is also used to: Establishing ontology concept information of the enterprise application domain, and determining ontology classes based on the ontology concept information; Acquire enterprise information data, obtain ontology-related data from the enterprise information data according to the ontology class, perform preprocessing, and obtain ontology knowledge; the ontology-related data includes time series information; A seed temporal knowledge graph is constructed based on the ontology knowledge.

Images