Identification analysis data governance and association method and system based on AI large model

Abstract

The invention provides an identification analysis data governance and association method and system based on an AI large model, and the method comprises the steps: carrying out the pre-classification of a new registration identification through an identification analysis registration module in the AI large model, and determining the industry type of the new registration identification; using a template industry attribute matching module in the AI large model to perform template classification on the new registration identifier, and determining a template category to which the new registration identifier belongs; performing identification analysis on the industry category and the template category by utilizing the inter-enterprise relationship knowledge graph, and mining enterprise information involved in identification analysis data; based on the preset association preconditions, causal association and process association between enterprise information are established, and through automatic identification analysis classification and identification, the association relationship between different identification analyses is automatically mined, so that the treatment level and value mining capability of identification analysis data are effectively improved; and powerful technical support is provided for intelligent development of an industrial internet identification analysis system.

Description

Technical Field

[0001] The present invention relates to the field of identity resolution technology, and in particular to an identity resolution data governance and association method and system based on an AI big model. Background Art

[0002] Identity resolution technology is key infrastructure for the Industrial Internet, serving as the internet's equivalent of the DNS system. By assigning a unique identifier to each physical or virtual object and establishing corresponding resolution services, accurate identification and location of these objects are achieved. This technology can resolve identifiers across industries, regions, and even countries, thereby facilitating data sharing and interaction between diverse systems. In industrial production, identity resolution technology enables precise tracking and management of products throughout their entire lifecycle, from raw material procurement to product sales, improving supply chain transparency and efficiency. Furthermore, identity resolution technology plays a vital role in the Internet of Things (IoT), providing identity recognition for a vast number of IoT devices, enabling interoperability and laying the foundation for building smart cities, smart homes, and other application scenarios.

[0003] However, existing identity resolution systems face the following challenges in data governance and correlation analysis: Low data classification efficiency: Traditional rule engines rely on manually pre-set labels, making them incapable of adapting to the dynamic classification needs of massive amounts of heterogeneous data. Insufficient correlation analysis depth: Data silos are severe between enterprises, lacking understanding of implicit relationships between identities. Poor industry adaptability: Identity registration templates are limited and fail to incorporate industry specificities, limiting the value of data utilization.

[0004] Therefore, how to improve the intelligence level of identity resolution data has become a technical problem that technical personnel in this field urgently need to solve. Summary of the Invention

[0005] The present invention provides an AI big model-based identity resolution data governance and association method and system to address the defect of poor intelligence level of identity resolution in the prior art.

[0006] In a first aspect, the present invention provides a method for managing and associating identity resolution data based on an AI big model, comprising: Use the identity resolution and registration module in the AI model to pre-classify the newly registered identity and determine the industry category to which the newly registered identity belongs; Use the template industry attribute matching module in the AI big model to classify the template of the newly registered logo and determine the template category to which the newly registered logo belongs; Performing identity resolution on the industry category and the template category using the inter-enterprise relationship knowledge graph, and mining enterprise information involved in the identity resolution data; Based on preset association preconditions, causal associations and process associations are established between the enterprise information.

[0007] According to the AI big model-based identity resolution data governance and association method provided by the present invention, before establishing the causal association and process association between the enterprise information, the method further includes: Determine the completeness and accuracy of data fields; Determine the timeliness, integrity, and privacy protection requirements of data; The integrity, accuracy, timeliness, completeness and privacy protection requirements are used as preset association preconditions.

[0008] According to an AI big model-based identifier resolution data governance and association method provided by the present invention, the identifier resolution registration module in the AI big model is used to pre-classify the newly registered identifier and determine the industry category to which the newly registered identifier belongs, including: Use the identity resolution and registration module in the AI model to identify industry characteristics; The industry category to which the new registered logo belongs is determined by combining the industry to which the industry feature belongs and the industry attributes of the associated secondary nodes.

[0009] According to an AI big model-based identifier resolution data governance and association method provided by the present invention, the template industry attribute matching module in the AI big model is used to classify the newly registered identifier and determine the template category to which the newly registered identifier belongs, including: Utilize the template industry attribute matching module in the AI big model to identify the data fields and format specifications of the newly registered logo; The data fields and the format specifications are compared with known industry template features to determine the template category to which the new registration logo belongs.

[0010] According to the AI big model-based identity resolution data governance and association method provided by the present invention, the method utilizes the inter-enterprise relationship knowledge graph to perform identity resolution on the industry category and the template category, and mines the enterprise information involved in the identity resolution data, including: Utilizing the inter-enterprise relationship knowledge graph to perform identity resolution on the industry category and the template category, and extract enterprise feature information; Identify the enterprise characteristic information and determine the cooperative relationship, supply chain relationship, equity relationship and strategic cooperation relationship between enterprises.

[0011] According to the AI big model-based identifier resolution data governance and association method provided by the present invention, before using the identifier resolution registration module in the AI big model to pre-classify the newly registered identifier, the method further includes: Collect sample data from the registration information database, parsing log database, and enterprise user database; Based on transfer learning, the pre-trained language model of the Transformer architecture is used to train the sample data with the professional knowledge and data characteristics in the field of identity resolution to build an AI large model.

[0012] According to the AI big model-based identity resolution data governance and association method provided by the present invention, after establishing the causal association and process association between the enterprise information, the method further includes: When the newly registered identifier is a newly emerged identifier data feature and industry dynamic changes, a new business attribute is automatically created.

[0013] According to the AI big model-based identity resolution data governance and association method provided by the present invention, after establishing the causal association and process association between the enterprise information, the method further includes: Receive user feedback and actual business results; Based on the user feedback and actual business results, the AI big model is optimized to improve data governance and correlation capabilities.

[0014] In a second aspect, the present invention further provides an identity resolution data governance and association system based on an AI big model, comprising: A data governance module is used to pre-classify the newly registered identifier using the identifier resolution registration module in the AI big model to determine the industry category to which the newly registered identifier belongs; and to perform template classification on the newly registered identifier using the template industry attribute matching module in the AI big model to determine the template category to which the newly registered identifier belongs; The data association module is used to use the inter-enterprise relationship knowledge graph to perform identity resolution on the industry category and the template category, and to mine the enterprise information involved in the identity resolution data; and to establish causal associations and process associations between the enterprise information based on preset association preconditions.

[0015] In a third aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the program, it implements any of the above-mentioned methods for identity resolution data governance and association based on the AI big model.

[0016] In a fourth aspect, the present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it implements any of the above-mentioned methods for identity resolution data governance and association based on the AI big model.

[0017] In a fifth aspect, the present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned methods for identity resolution data governance and association based on the AI big model.

[0018] The AI big model-based identity resolution data governance and association method and system provided by the present invention include: using the identity resolution registration module in the AI big model to pre-classify the newly registered identity and determine the industry category to which the newly registered identity belongs; using the template industry attribute matching module in the AI big model to perform template classification on the newly registered identity and determine the template category to which the newly registered identity belongs; using the inter-enterprise relationship knowledge graph to perform identity resolution on industry categories and template categories, and mine the enterprise information involved in the identity resolution data; based on preset association preconditions, establish causal associations and process associations between enterprise information, through automatic identity resolution classification and identification, and automatically mine the association relationship between different identity resolutions, effectively improve the governance level and value mining capabilities of identity resolution data, provide strong technical support for the intelligent development of the industrial Internet identity resolution system, and have broad application prospects and market potential. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 This is a flowchart of the method for managing and associating identity resolution data based on the AI big model provided in this embodiment; Figure 2 This is a schematic diagram of the structure of the identity resolution data governance and association system based on the AI big model provided in this embodiment; Figure 3 Schematic diagram of the structure of the electronic device provided in this embodiment. DETAILED DESCRIPTION

[0021] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0022] Figure 1 This is a flow chart of the identity resolution data governance and association method based on the AI big model provided in this embodiment.

[0023] like Figure 1As shown, the method for managing and associating identity resolution data based on an AI big model provided by an embodiment of the present invention can be executed by an identity resolution system and mainly includes the following steps: 101. Use the identity resolution and registration module in the AI big model to pre-classify the newly registered identity and determine the industry category to which the newly registered identity belongs.

[0024] In a specific implementation process, the identity resolution system, as the core infrastructure connecting physical objects and digital information, is widely used in supply chain management, product traceability, equipment operation and maintenance and other fields. The identity resolution system in this embodiment uses the semantic understanding and knowledge reasoning capabilities of the AI big model (Artificial Intelligence) to build an industry attribute-driven identity data governance framework and a cross-system association analysis engine to efficiently implement data resolution, etc. It mainly includes an intelligent classification module: based on the industry knowledge base and big model semantic analysis, it realizes the automatic classification and label generation of identifiers. Deep association module: through the big model, it mines the explicit / implicit relationships between identifiers and builds a cross-enterprise and cross-system data association network.

[0025] Among them, the identity resolution and registration module based on industry attributes is mainly used to introduce industry attribute labels during the identity resolution and registration stage. By pre-classifying the identities of different industries and using the AI big model to learn and understand the industry characteristics, the big model automatically assigns the industry category to the newly registered identity based on the industry selected by the enterprise at the time of registration and the industry attributes of the associated secondary nodes. Identifiers that need to be resolved and associated are defined as newly registered identifiers. Therefore, by identifying their industry characteristics through the identity resolution and registration module, and combining them with the industry attributes of the industry and the associated secondary nodes, the industry category to which the newly registered identity belongs can be quickly determined. For example, in the automobile manufacturing industry, identifiers can be subdivided into subcategories such as parts production identifiers and vehicle assembly identifiers; in the electronic information industry, they can be divided into chip identifiers, electronic product assembly identifiers, etc. Therefore, preliminary classification is achieved at the source of the data, laying the foundation for subsequent data governance.

[0026] The logo resolution and registration module in the AI model can be used to pre-classify the newly registered logo and determine the industry category to which the newly registered logo belongs. The following methods can also be used: First, a heterogeneous data cleaning engine normalizes the input of newly registered identifiers (including multimodal identifier information such as text, images, and audio). A generative adversarial network is used to remove noise and retain key semantic features. Second, a large-scale multidimensional encoder (such as a hybrid architecture of RoBERTa-wwm and CLIP) is used to extract deep semantic vectors from text and topological features from visual symbols, respectively. A cross-modal attention mechanism is then used to align the feature spaces. Subsequently, the system integrates with a dynamic industry knowledge graph. By crawling business data, relevant literature, and industry white papers in real time, a graph neural network is used to construct an industry concept tree with timestamp attributes. A subgraph matching algorithm is then used to calculate the multidimensional similarity between the features of the newly registered identifier and industry nodes. A cascaded classification architecture is employed in the decision-making stage. The first layer uses comparative learning to select the top-K candidate industries. The second layer uses an interpretable classifier (such as a prototype network) combined with association rules in the knowledge graph for fine-grained verification. Finally, the industry category of the newly registered identifier is output, along with a confidence score and traceability path. An incremental learning mechanism is then initiated to automatically convert edge cases into training samples for model optimization. This method breaks through the limitations of single-modal classification, dynamically perceives industry evolution through time and space dimensions, and effectively improves the classification accuracy in emerging e-commerce industries and interdisciplinary fields.

[0027] 102. Use the template industry attribute matching module in the AI big model to classify the templates of the newly registered logo and determine the template category to which the newly registered logo belongs.

[0028] Specifically, the template industry attribute matching module can be used, which is mainly used to assign industry attributes to the identification resolution template. The AI big model compares the data fields, format specifications and other information contained in the template with the known industry template features to determine the industry to which it belongs. For example, a template containing fields such as product size, material, and production process parameters, the big model can determine that it belongs to the machinery manufacturing industry template. By matching the industry attributes of the template, the classification granularity of the identification resolution data is further refined to ensure that the identification data in the same industry follows similar template specifications, which is convenient for subsequent data integration and analysis.

[0029] Specifically, the template category of newly registered logos can also be identified based on an AI template classification system that is enhanced with adversarial templates and evolution tracking. This mainly includes: first, building an industry-adaptive template library, inheriting the industry node features of the dynamic knowledge graph, and using the graph attention network (GAT) to mine the meta-structural features of templates in various industries (including text keyword distribution, visual symbol space topology, and audio spectrum patterns), generating template embedding vectors with industry relevance. Secondly, in the feature matching stage, an innovative hybrid contrastive learning framework is designed to bidirectionally map the multimodal features of the new logo encoded by the large model (reusing the RoBERTa-wwm-CLIP hybrid encoder) to the template library. The cosine similarity with the standard template is calculated through supervised contrastive learning, and the potential correlation pattern of cross-industry templates is captured through self-supervised comparison. In particular, an adversarial template generator is introduced to dynamically create boundary samples to enhance the model's ability to identify emerging templates. Finally, a template conflict resolution mechanism was employed. By constructing a Bayesian probabilistic graphical model, the model integrates template matching, industry association weights (derived from previous classification results), and the strength of semantic associations within the knowledge graph. Multiple rounds of game-based decision-making were initiated for candidate templates whose overlap exceeded a threshold. Attention weight visualization techniques were then incorporated to ensure interpretability of the template classification results. A template evolution tracking algorithm was employed, utilizing a temporal convolutional network (TCN) to analyze template morphological changes within the industrial and commercial registration data stream, enabling weekly dynamic updates of the template feature space. This effectively improved the accuracy of template category recognition.

[0030] As business evolves and data volumes increase, the AI big model continuously learns emerging identification data characteristics and industry dynamics. When it detects changes in the business attributes of certain identification data or the emergence of new industry segments, it automatically and dynamically adjusts data classifications. For example, when an emerging smart hardware product emerges whose identification data combines the characteristics of traditional electronic products with unique IoT connectivity properties, the big model can promptly create new classification categories to incorporate this identification data and update relevant data governance policies.

[0031] 103. Use the knowledge graph of inter-enterprise relationships to perform identity resolution on industry categories and template categories, and mine the enterprise information involved in the identity resolution data.

[0032] Specifically, the knowledge graph of inter-enterprise relationships can be used to perform identity resolution on industry categories and template categories, extract enterprise feature information, identify enterprise feature information, and determine the cooperative relationships, supply chain relationships, equity relationships, and strategic cooperation relationships between enterprises. Utilize the powerful knowledge graph construction capabilities of the AI big model to analyze the cooperative relationships, supply chain relationships, etc. between enterprises. By extracting and associating the enterprise information involved in the identity resolution data, for example, if an automobile manufacturer frequently interacts with the identification data of multiple parts suppliers, the big model can mine the supply relationships between them and establish corresponding nodes and edges in the knowledge graph. Furthermore, non-identification data information such as equity relationships and strategic cooperation agreements between enterprises can be analyzed to enrich the dimensions of inter-enterprise relationships and provide more comprehensive background support for the association analysis of identity resolution data.

[0033] Specifically, the enterprise information may be mined by an identified enterprise information mining system based on hyper-relational graph reasoning and joint optimization, including: First, we construct a hyper-relational knowledge graph of enterprises, integrating industrial and commercial registration information (equity structure, business scope), supply chain data (upstream and downstream cooperation) and investment relations (M&A events), and use graph embedding technology to encode enterprise nodes into multi-dimensional vectors containing industry attributes and template features.

[0034] Secondly, a dual-channel graph attention network is designed. The main channel uses enterprise-industry-template triples for cross-layer propagation to capture enterprise association patterns implicit in identity resolution data (such as when enterprises in the same industry share specific template combinations). The auxiliary channel uses a spatiotemporal graph convolutional network (ST-GCN) to analyze the temporal evolution of enterprise relationships. During the inference phase, a dynamic pruning strategy is used to sample subgraphs of the graph based on the confidence of the previous classification. An improved label propagation algorithm is used to map identity features to enterprise nodes, and multi-hop reasoning is used to identify clusters of potentially related enterprises.

[0035] Finally, a joint optimization mechanism is activated to reverse-inject the enterprise relationship inference results into the industry classification module. The graph neural network updates the weights of industry nodes in the dynamic knowledge graph, while simultaneously triggering the adversarial enhancement module in the template library to generate targeted training samples, effectively mining enterprise information. By combining the enterprise relationship influence evaluation matrix with the PageRank algorithm and semantic similarity calculation, a visual enterprise network graph with traceable association paths is output, effectively improving the accuracy of enterprise relationship mining.

[0036] 104. Based on preset association preconditions, establish causal relationships and process relationships between enterprise information.

[0037] Under certain preconditions, deep associations between different identification resolutions can also be achieved. Taking the entire product life cycle as an example, from raw material procurement identification, production and processing identification to product sales identification, the AI big model analyzes the time series, business process sequence and other information of these identification data to establish causal associations and process associations between them. For example, the batch number in the raw material procurement identification matches the raw material batch field in the production and processing identification, and the product model in the production and processing identification is consistent with the product model in the sales identification. Based on this, the big model can construct a complete identification chain for the product from raw materials to finished products, realize the seamless connection and deep integration of different identification resolution data, and provide accurate data support for the company's production management, quality traceability and other businesses.

[0038] In order to ensure the accuracy and reliability of identity resolution data association, a series of preconditions have been set. First, data integrity and accuracy are the foundation, requiring the identity resolution system to strictly follow data quality standards during data collection and entry to ensure that data fields are not missing and that the data content is authentic and reliable. Secondly, the timeliness of the data is also crucial. For some time-sensitive identification data, such as traceability identification of fresh products and industrial production identification with strong timeliness, meaningful results can only be obtained by performing association analysis within the specified time frame. In addition, data security and privacy protection requirements must also be considered. When associating data between enterprises, ensure that data transmission and storage comply with relevant security standards, and encrypt data involving corporate commercial secrets and personal privacy. Only when these preconditions are met can the AI large model fully exert its data association analysis capabilities and provide valuable association results for the identity resolution system.

[0039] Furthermore, based on the above embodiment, this embodiment also includes collecting sample data from the registration information library, parsing log library, and enterprise user database. This collected data is pre-processed by cleaning, deduplication, format conversion, and other operations to ensure that the data quality meets the input requirements of the large AI model. For example, invalid records in the parsing log are removed, and the data format of different enterprise user databases is unified.

[0040] Select a suitable large-scale AI model architecture, such as a pre-trained language model using the Transformer architecture, and fine-tune it based on domain expertise and data characteristics. During training, use a large amount of historical identity resolution data and industry knowledge text as training data. By adjusting model hyperparameters and optimizing algorithms, improve the model's data classification accuracy and association analysis capabilities. For example, employ transfer learning techniques to transfer knowledge from a pre-trained model in general domains to the identity resolution domain, followed by targeted training and optimization.

[0041] The pre-processed identity resolution data is input into the trained AI big model for real-time data classification and association analysis. Based on the model output, the data classification catalog and association relationship map of the identity resolution system are updated. For example, for newly registered identities, the model automatically classifies them and assigns them to the corresponding industry category and template category. For data in the parsed log, the model mines the associations and identity links between enterprises and feeds this information back to the system to provide data support for the enterprise's business decision-making. At the same time, a feedback mechanism for data classification and association results is established. Based on user feedback and actual business results, the AI big model is continuously optimized to continuously improve the performance of data governance and association analysis.

[0042] Based on the same general inventive concept, the present invention also protects an identity resolution data governance and association system based on an AI big model. The following description of the XX device provided by the present invention can be referred to in correspondence with the above description of the XX device provided by the present invention.

[0043] Figure 2 This is a structural diagram of the identity resolution data governance and association system based on the AI big model provided in this embodiment.

[0044] like Figure 2 As shown, this embodiment provides an identity resolution data governance and association system based on an AI big model, including: The data governance module 201 is configured to pre-classify the newly registered identifier using the identifier resolution registration module in the AI big model to determine the industry category to which the newly registered identifier belongs; and to perform template classification on the newly registered identifier using the template industry attribute matching module in the AI big model to determine the template category to which the newly registered identifier belongs; The data association module 202 is used to use the inter-enterprise relationship knowledge graph to perform identity resolution on the industry category and the template category, and to mine the enterprise information involved in the identity resolution data; and to establish causal associations and process associations between the enterprise information based on preset association preconditions.

[0045] Figure 3 Schematic diagram of the structure of the electronic device provided in this embodiment.

[0046] like Figure 3As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other via the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute an identity resolution data governance and association method based on an AI big model, the method comprising: using the identity resolution registration module in the AI big model to pre-classify the newly registered identity and determine the industry category to which the newly registered identity belongs; using the template industry attribute matching module in the AI big model to perform template classification on the newly registered identity and determine the template category to which the newly registered identity belongs; using the inter-enterprise relationship knowledge graph to perform identity resolution on the industry category and the template category, and mine the enterprise information involved in the identity resolution data; and establishing causal associations and process associations between the enterprise information based on preset association preconditions.

[0047] Furthermore, the logic instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0048] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the identifier resolution data governance and association method based on the AI big model provided by the above methods. The method includes: using the identifier resolution registration module in the AI big model to pre-classify the newly registered identifier, and determine the industry category to which the newly registered identifier belongs; using the template industry attribute matching module in the AI big model to perform template classification on the newly registered identifier, and determine the template category to which the newly registered identifier belongs; using the inter-enterprise relationship knowledge graph to perform identifier resolution on the industry category and the template category, and mine the enterprise information involved in the identifier resolution data; based on preset association preconditions, establish causal associations and process associations between the enterprise information.

[0049] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the identifier resolution data governance and association method based on the AI big model provided by the above-mentioned methods, the method comprising: using the identifier resolution registration module in the AI big model to pre-classify the newly registered identifier, and determine the industry category to which the newly registered identifier belongs; using the template industry attribute matching module in the AI big model to perform template classification on the newly registered identifier, and determine the template category to which the newly registered identifier belongs; using the inter-enterprise relationship knowledge graph to perform identifier resolution on the industry category and the template category, and mine the enterprise information involved in the identifier resolution data; based on preset association preconditions, establish causal associations and process associations between the enterprise information.

[0050] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0051] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0052] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for managing and associating identity resolution data based on an AI big model, characterized in that: include: Use the identity resolution and registration module in the AI model to pre-classify the newly registered identity and determine the industry category to which the newly registered identity belongs; Use the template industry attribute matching module in the AI big model to classify the template of the newly registered logo and determine the template category to which the newly registered logo belongs; Performing identity resolution on the industry category and the template category using the inter-enterprise relationship knowledge graph, and mining enterprise information involved in the identity resolution data; Based on preset association preconditions, causal associations and process associations are established between the enterprise information.

2. The method for managing and associating identity resolution data based on an AI big model according to claim 1 is characterized in that: Before establishing the causal relationship and process relationship between the enterprise information, the method further includes: Determine the completeness and accuracy of data fields; Determine the timeliness, integrity, and privacy protection requirements of data; The integrity, accuracy, timeliness, completeness and privacy protection requirements are used as preset association preconditions.

3. The method for managing and associating identity resolution data based on an AI big model according to claim 1 is characterized in that: The identification resolution and registration module in the AI large model is used to pre-classify the newly registered identification and determine the industry category to which the newly registered identification belongs, including: Use the identity resolution and registration module in the AI model to identify industry characteristics; The industry category to which the new registered logo belongs is determined by combining the industry to which the industry feature belongs and the industry attributes of the associated secondary nodes.

4. The method for managing and associating identity resolution data based on an AI big model according to claim 1 is characterized in that: The template industry attribute matching module in the AI big model is used to classify the new registration logo into templates to determine the template category to which the new registration logo belongs, including: Utilize the template industry attribute matching module in the AI big model to identify the data fields and format specifications of the newly registered logo; The data fields and the format specifications are compared with known industry template features to determine the template category to which the new registration logo belongs.

5. The method for managing and associating identity resolution data based on an AI big model according to claim 1 is characterized in that: The method of performing identity resolution on the industry category and the template category by utilizing the inter-enterprise relationship knowledge graph and mining the enterprise information involved in the identity resolution data includes: Utilizing the inter-enterprise relationship knowledge graph to perform identity resolution on the industry category and the template category, and extract enterprise feature information; Identify the enterprise characteristic information and determine the cooperative relationship, supply chain relationship, equity relationship and strategic cooperation relationship between enterprises.

6. The method for managing and associating identity resolution data based on an AI big model according to claim 1 is characterized in that: Before the identification resolution registration module in the AI large model is used to pre-classify the newly registered identification, the method further includes: Collect sample data from the registration information database, parsing log database, and enterprise user database; Based on transfer learning, the pre-trained language model of the Transformer architecture is used to train the sample data with the professional knowledge and data characteristics in the field of identity resolution to build an AI large model.

7. The method for managing and associating identity resolution data based on an AI big model according to any one of claims 1 to 6, characterized in that: After establishing the causal relationship and process relationship between the enterprise information, the method further includes: When the newly registered identifier is a newly emerged identifier data feature and industry dynamic changes, a new business attribute is automatically created.

8. The method for managing and associating identity resolution data based on an AI big model according to any one of claims 1 to 6, characterized in that: After establishing the causal relationship and process relationship between the enterprise information, the method further includes: Receive user feedback and actual business results; Based on the user feedback and actual business results, the AI big model is optimized to improve data governance and correlation capabilities.

9. An identity resolution data management and association system based on an AI big model, characterized by: include: A data governance module is used to pre-classify newly registered identifiers using the identifier resolution and registration module in the AI large model to determine the industry category to which the newly registered identifiers belong; Use the template industry attribute matching module in the AI big model to classify the template of the newly registered logo and determine the template category to which the newly registered logo belongs; A data association module is used to perform identity resolution on the industry category and the template category using the inter-enterprise relationship knowledge graph, and to mine enterprise information involved in the identity resolution data; It is used to establish causal associations and process associations between the enterprise information based on preset association preconditions.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, it implements the identity resolution data governance and association method based on the AI big model as described in any one of claims 1 to 8.

Images