Associated enterprise identification method and device, equipment and medium

Abstract

The embodiment of the invention discloses an associated enterprise identification method and device, equipment and a medium, and the method comprises the steps: obtaining the associated identification information of a specified enterprise, the associated identification information being a plurality of pieces of preset identity type information; respectively matching the associated identification information with a plurality of pieces of identity type information in pre-screened associated identification information of a plurality of enterprises to obtain a matching result of each piece of identity type information; determining association indexes of the plurality of enterprises and the specified enterprise based on matching results of the identity type information; and identifying the enterprises of which the association indexes are greater than a preset threshold value as associated enterprises of the specified enterprise. By presetting a plurality of identity type information and matching the identity type information with the pre-screened enterprises, the associated enterprises can be identified more comprehensively and accurately, and the situations of misjudgment and missed judgment are reduced.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and medium for identifying an associated enterprise. Background Art

[0002] In the context of economic globalization and increasingly fierce market competition, the relationships between enterprises are becoming increasingly complex and hidden. When banks and related units conduct risk assessment and management on enterprises, they often face various risks and even suffer actual losses due to insufficient identification of related enterprises.

[0003] At present, collecting data from enterprises and individual customers from a variety of public sources, integrating, analyzing and mining them to analyze the relationships between enterprises has become a key issue in financial risk prevention and control. However, existing methods and technologies still face many challenges in processing large-scale data, identifying complex relationships, and ensuring the legitimacy and compliance of data. Summary of the invention

[0004] One or more embodiments of the present specification provide a method, device, equipment and medium for identifying an associated enterprise, which are used to solve the technical problems raised by the background technology.

[0005] One or more embodiments of this specification adopt the following technical solutions:

[0006] One or more embodiments of this specification provide a method for identifying an associated enterprise, the method comprising:

[0007] Obtaining the associated identification information of the specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information;

[0008] Matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, respectively, to obtain matching results of each identity type information;

[0009] Determining, based on the matching results of each identity type information, the association index between the multiple enterprises and the designated enterprise;

[0010] The enterprises whose association index is greater than a preset threshold are identified as associated enterprises of the designated enterprise.

[0011] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0012] Improve the accuracy of related enterprise identification: By pre-setting multiple identity type information and matching them with pre-screened enterprises, related enterprises can be identified more comprehensively and accurately, reducing misjudgments and missed judgments.

[0013] Enhance risk assessment and management capabilities: Help banks and related units better understand the relationships between enterprises, so as to more accurately assess risks, formulate corresponding risk management strategies and reduce potential risks.

[0014] Improve data processing efficiency: Able to effectively process large-scale data, quickly match and analyze related identification information, improve work efficiency, and save time and resources.

[0015] Adapting to complex relationships: This method can better identify and analyze increasingly complex and hidden enterprise relationships, providing deeper insights.

[0016] Ensure data legality and compliance: When collecting and processing data, follow the principles of legality and compliance to protect the privacy and data security of enterprises and individual customers.

[0017] Improving the level of financial risk prevention and control: Providing more reliable risk prevention and control support to financial institutions will help maintain the stability and security of the financial market.

[0018] Furthermore, before matching the associated identification information with the plurality of identity type information in the associated identification information of the plurality of pre-screened enterprises, the method further includes:

[0019] Obtaining interaction data between the designated enterprise and other enterprises through multiple channels;

[0020] A plurality of enterprises that interact with the designated enterprise are screened out from the interaction data.

[0021] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0022] Improve the accuracy of identifying related companies: By obtaining interaction data from multiple channels and screening out multiple companies that interact with a specified company, you can have a more comprehensive understanding of the business relationships of the specified company, thereby identifying related companies more accurately.

[0023] Enhanced risk assessment and management capabilities: The interactive data obtained can provide detailed information on business transactions between enterprises, helping banks and related units to more deeply assess risks and develop more effective risk management strategies.

[0024] Expanding the coverage of related enterprises: Obtaining interaction data through multiple channels can discover some potential related enterprises that may be overlooked in traditional association identification methods, thereby expanding the coverage of related enterprises.

[0025] Providing a more comprehensive corporate portrait: Interaction data can supplement the associated identification information of a specified enterprise, providing a more comprehensive corporate portrait and helping banks and related units better understand the business model and operation of the enterprise.

[0026] Discover new relationships in a timely manner: As the business of an enterprise develops and changes, new relationships may continue to emerge. By acquiring and screening interactive data in real time, these new relationships can be discovered in a timely manner, providing timely support for risk prevention and control.

[0027] Improve the credibility and reliability of data: Data obtained from multiple channels can verify and supplement each other, improve the credibility and reliability of data, and thus enhance the accuracy of identifying related companies.

[0028] Furthermore, the multiple identity type information includes one or more of business information, annual report information, domain name information and intellectual property information.

[0029] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0030] Comprehensive understanding of the enterprise: covering industrial and commercial information, annual report information, domain name information and intellectual property information, etc., can provide a more comprehensive understanding of the basic situation, operating conditions, network activities and intellectual property status of the enterprise.

[0031] Improve the accuracy of identifying related companies: Different types of information complement and verify each other, which helps to more accurately identify related companies and reduce the possibility of misjudgment and omission.

[0032] Gain insights into corporate relationships: This information can provide clues about potential connections and relationships between companies, helping to discover hidden affiliated companies.

[0033] Furthermore, the associated identification information is matched with multiple identity type information in the associated identification information of multiple enterprises screened in advance, and matching results of each identity type information are obtained, including:

[0034] Each piece of identity type information of the designated enterprise is compared with the corresponding identity type information of the multiple enterprises in turn to obtain the similarity between each piece of identity type information of the designated enterprise and the corresponding identity type information of each enterprise.

[0035] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0036] More accurate identification of related companies: By comparing the similarity of each identity type information, the degree of association between companies can be determined more accurately, improving the accuracy of related company identification.

[0037] Comprehensively evaluate association relationships: Taking into account multiple identity type information, it is possible to evaluate the relationship between enterprises from different angles and provide a more comprehensive association analysis.

[0038] Discover potential associations: Similarity comparisons can reveal some potential associations, even if companies are not exactly the same in some aspects, they may be similar in other aspects.

[0039] Furthermore, before matching the associated identification information with the plurality of identity type information in the associated identification information of the plurality of pre-screened enterprises, the method further includes:

[0040] Social network analysis technology is used to analyze the preset relationship network between the designated enterprise and other enterprises, and multiple enterprises associated with the designated enterprise are screened out. The preset relationship network includes investment relationships, cooperation relationships, and supply chain relationships between the designated enterprise and other enterprises.

[0041] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0042] More comprehensive identification of related companies: Social network analysis technology can be used to discover relationships between companies that may be overlooked by traditional methods, such as indirect investments, cooperative alliances, etc., thereby more comprehensively identifying related companies.

[0043] Gain in-depth understanding of the enterprise ecosystem: By analyzing the preset relationship network, including investment relationships, partnerships, and supply chain relationships, we can gain a deeper understanding of the business ecosystem in which a given enterprise is located, as well as the interactions and dependencies between it and other enterprises.

[0044] Improve the accuracy of risk assessment: A comprehensive understanding of the relationships between enterprises helps to more accurately assess risks, as there may be risk transmission and mutual influence between related enterprises. This can help banks and related units better formulate risk management strategies.

[0045] Furthermore, before matching the associated identification information with the plurality of identity type information in the associated identification information of the plurality of pre-screened enterprises, the method further includes:

[0046] Use machine learning algorithms to train and learn the associated identification information of enterprises, and obtain a neural network model for screening the associated enterprises of an enterprise;

[0047] Based on the neural network model, multiple enterprises associated with the designated enterprise are obtained by screening.

[0048] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0049] Improved screening accuracy: Machine learning algorithms can automatically learn and identify patterns and features in enterprise-related identification information, thereby more accurately screening multiple enterprises associated with a given enterprise.

[0050] Adapting to complex relationships: The neural network model can handle complex enterprise relationship networks, including nonlinear relationships and multi-factor influences, and improve the ability to identify related enterprises.

[0051] Dynamically adapt to changes: By continuously updating training data, machine learning algorithms can adapt to the dynamic changes in corporate relationships and capture new associations in a timely manner.

[0052] One or more embodiments of this specification provide a device for identifying an associated enterprise, including:

[0053] An identification information acquisition unit, which acquires the associated identification information of a specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information;

[0054] A matching unit, matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, to obtain matching results of each identity type information;

[0055] an association index determination unit, which determines an association index between the plurality of enterprises and the designated enterprise based on a matching result of each identity type information;

[0056] The associated enterprise identification unit identifies the enterprises whose association index is greater than a preset threshold as associated enterprises of the designated enterprise.

[0057] Furthermore, before matching the associated identification information with the plurality of identity type information in the associated identification information of the plurality of pre-screened enterprises, the device further comprises:

[0058] An interaction data acquisition unit, which acquires interaction data between the designated enterprise and other enterprises through multiple channels;

[0059] The interactive enterprise screening unit screens out a plurality of enterprises that interact with the designated enterprise from the interactive data.

[0060] One or more embodiments of this specification provide an associated enterprise identification device, including:

[0061] at least one processor; and,

[0062] a memory communicatively connected to the at least one processor; wherein,

[0063] The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0064] Obtaining the associated identification information of the specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information;

[0065] Matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, respectively, to obtain matching results of each identity type information;

[0066] Determining, based on the matching results of each identity type information, the association index between the multiple enterprises and the designated enterprise;

[0067] The enterprises whose association index is greater than a preset threshold are identified as associated enterprises of the designated enterprise.

[0068] One or more embodiments of this specification provide a non-volatile computer storage medium storing computer executable instructions, which can achieve the following when executed by a computer:

[0069] Obtaining the associated identification information of the specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information;

[0070] Matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, respectively, to obtain matching results of each identity type information;

[0071] Determining, based on the matching results of each identity type information, the association index between the multiple enterprises and the designated enterprise;

[0072] The enterprises whose association index is greater than a preset threshold are identified as associated enterprises of the designated enterprise.

[0073] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects:

[0074] Improve the accuracy of related enterprise identification: By pre-setting multiple identity type information and matching them with pre-screened enterprises, related enterprises can be identified more comprehensively and accurately, reducing misjudgments and missed judgments.

[0075] Enhance risk assessment and management capabilities: Help banks and related units better understand the relationships between enterprises, so as to more accurately assess risks, formulate corresponding risk management strategies and reduce potential risks.

[0076] Improve data processing efficiency: Able to effectively process large-scale data, quickly match and analyze related identification information, improve work efficiency, and save time and resources.

[0077] Adapting to complex relationships: This method can better identify and analyze increasingly complex and hidden enterprise relationships, providing deeper insights.

[0078] Ensure data legality and compliance: When collecting and processing data, follow the principles of legality and compliance to protect the privacy and data security of enterprises and individual customers.

[0079] Improving the level of financial risk prevention and control: Providing more reliable risk prevention and control support to financial institutions will help maintain the stability and security of the financial market. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art description. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. In the drawings:

[0081] Figure 1 A flowchart of a method for identifying an associated enterprise provided for one or more embodiments of this specification;

[0082] Figure 2 A schematic diagram of association level determination provided for one or more embodiments of this specification;

[0083] Figure 3 A schematic diagram of the structure of an associated enterprise identification device provided in one or more embodiments of this specification;

[0084] Figure 4 A schematic diagram of the structure of an associated enterprise identification device provided in one or more embodiments of this specification. DETAILED DESCRIPTION

[0085] The embodiments of this specification provide a method, device, equipment and medium for identifying an associated enterprise.

[0086] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0087] Figure 1This is a flowchart of a method for identifying an associated enterprise provided in one or more embodiments of this specification, and the process can be executed by an associated enterprise identification system. Certain input parameters or intermediate results in the process allow manual intervention and adjustment to help improve accuracy.

[0088] The method steps of the embodiment of this specification are as follows:

[0089] S102, obtaining the associated identification information of the designated enterprise, wherein the associated identification information is a plurality of pre-set identity type information.

[0090] In the embodiment of the present specification, the multiple identity type information may include one or more of business information, annual report information, domain name information and intellectual property information.

[0091] It should be noted that the above content can be implemented through the following specific implementation plans:

[0092] Data collection: Determine the type of identity information that needs to be obtained, such as industrial and commercial information, annual report information, domain name information, and intellectual property information. Collect relevant data sources, such as departments, corporate credit information disclosure systems, annual report databases, domain name registration agencies, intellectual property offices, etc. Establish data collection channels, and obtain data through interface calls, data crawling, data purchase, etc.

[0093] Data cleaning and organization: Clean and pre-process the collected data to remove noise, duplicate data, and erroneous information. Convert and standardize data formats to ensure consistency and comparability of data from different sources. Organize and classify data to store and manage information of different identity types separately.

[0094] Data association and integration: Establish associations between corporate identities and information of different identity types, such as through corporate names, unified social credit codes, etc. Integrate information of multiple identity types into one data set for comprehensive analysis and application.

[0095] Data storage and management: Choose appropriate data storage methods, such as databases, data warehouses, or data lakes, to ensure data security and accessibility. Establish data management mechanisms, including data backup, updates, and maintenance, to ensure data accuracy and integrity.

[0096] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0097] Comprehensive understanding of the enterprise: covering industrial and commercial information, annual report information, domain name information and intellectual property information, etc., can provide a more comprehensive understanding of the basic situation, operating conditions, network activities and intellectual property status of the enterprise.

[0098] Improve the accuracy of identifying related companies: Different types of information complement and verify each other, which helps to more accurately identify related companies and reduce the possibility of misjudgment and omission.

[0099] Gain insights into corporate relationships: This information can provide clues about potential connections and relationships between companies, helping to discover hidden affiliated companies.

[0100] S104, matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, and obtaining matching results of each identity type information.

[0101] In the embodiment of the present specification, each identity type information of the designated enterprise may be compared with the corresponding identity type information of the multiple enterprises in turn to obtain the similarity between each identity type information of the designated enterprise and the corresponding identity type information of each enterprise.

[0102] It should be noted that the above content can be implemented through the following specific implementation plans:

[0103] Data preparation: Determine the identity type information that needs to be compared, such as business information, annual report information, domain name information, and intellectual property information. Collect the corresponding identity type information of the specified enterprise and multiple enterprises, and organize it into a comparable data format.

[0104] Select similarity calculation method: Select an appropriate similarity calculation method based on the characteristics of the identity type information. For example, for text information, you can use text similarity algorithms such as cosine similarity and Jaccard similarity; for numerical information, you can use distance measurement methods such as Euclidean distance and Manhattan distance.

[0105] Perform similarity calculation: for each identity type information of the specified enterprise, compare it with the corresponding identity type information of multiple enterprises in turn. Use the selected similarity calculation method to calculate the similarity between the identity type information of the specified enterprise and the identity type information corresponding to each enterprise.

[0106] Result storage and analysis: The calculated similarity results can be stored in a database, data warehouse or other suitable data storage methods. The similarity results can be analyzed to sort the enterprises according to the similarity level and find the enterprises with the highest similarity to the specified enterprise.

[0107] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0108] More accurate identification of related companies: By comparing the similarity of each identity type information, the degree of association between companies can be determined more accurately, improving the accuracy of related company identification.

[0109] Comprehensively evaluate association relationships: Taking into account multiple identity type information, it is possible to evaluate the relationship between enterprises from different angles and provide a more comprehensive association analysis.

[0110] Discover potential associations: Similarity comparisons can reveal some potential associations, even if companies are not exactly the same in some aspects, they may be similar in other aspects.

[0111] Furthermore, before matching the associated identification information with multiple identity type information in the associated identification information of multiple pre-screened enterprises, interaction data between the designated enterprise and other enterprises can be obtained through multiple channels; and multiple enterprises that interact with the designated enterprise can be screened out from the interaction data.

[0112] It should be noted that the above content can be implemented through the following specific implementation plans:

[0113] Identify the following data sources:

[0114] Online business database: Access professional business databases to obtain basic information, financial data, business scope, etc. of enterprises.

[0115] Public data: query the enterprise registration information, tax data, regulatory records, etc. released by the department.

[0116] Social media and news platforms: Monitor the company's official accounts, news reports, industry forums, etc. on social media to obtain the company's dynamics and interactive information.

[0117] Official website of the company: Browse the official website of the company to learn about its products, services, partners and other information.

[0118] Collect interactive data: Use web crawler technology or data collection tools to capture relevant data from selected data source channels. Clean and pre-process the collected data to remove noise and duplicate information to ensure data accuracy and consistency.

[0119] Screening of enterprises with interactions: Analyze interaction data and extract interaction behaviors related to the specified enterprise, such as contract signing, cooperation projects, transaction records, etc. Based on the characteristics and patterns of interaction behaviors, identify other enterprises that interact with the specified enterprise. Data mining, machine learning or statistical analysis methods can be used to assist the screening process and improve accuracy and efficiency.

[0120] Verification and supplementary information: Further verification of the selected enterprises to ensure that they do have an interactive relationship with the designated enterprises. More information about these enterprises, such as corporate background, industry status, reputation, etc., is obtained through other channels to enrich the understanding of them.

[0121] Establish an associated identification information database: organize the selected enterprises and their associated identification information into a database or data warehouse. Regularly update and maintain the associated identification information database to reflect changes in the relationships between enterprises.

[0122] Through the above implementation plan, the interaction data between the designated enterprise and other enterprises can be obtained through multiple channels, and multiple enterprises with interactions can be screened out to provide a basis for subsequent matching of related identification information. This can provide a more comprehensive understanding of the business network and related relationships of the enterprise, which is helpful for risk assessment, market analysis, business decision-making and other activities. At the same time, attention should be paid to the legality, security and privacy protection of the data to ensure compliance with relevant laws and regulations during data collection and use.

[0123] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0124] Improve the accuracy of identifying related companies: By obtaining interaction data from multiple channels and screening out multiple companies that interact with a specified company, you can have a more comprehensive understanding of the business relationships of the specified company, thereby identifying related companies more accurately.

[0125] Enhanced risk assessment and management capabilities: The interactive data obtained can provide detailed information on business transactions between enterprises, helping banks and related units to more deeply assess risks and develop more effective risk management strategies.

[0126] Expanding the coverage of related enterprises: Obtaining interaction data through multiple channels can discover some potential related enterprises that may be overlooked in traditional association identification methods, thereby expanding the coverage of related enterprises.

[0127] Providing a more comprehensive corporate portrait: Interaction data can supplement the associated identification information of a specified enterprise, providing a more comprehensive corporate portrait and helping banks and related units better understand the business model and operation of the enterprise.

[0128] Discover new relationships in a timely manner: As the business of an enterprise develops and changes, new relationships may continue to emerge. By acquiring and screening interactive data in real time, these new relationships can be discovered in a timely manner, providing timely support for risk prevention and control.

[0129] Improve the credibility and reliability of data: Data obtained from multiple channels can verify and supplement each other, improve the credibility and reliability of data, and thus enhance the accuracy of identifying related companies.

[0130] S106: Determine the association index between the multiple enterprises and the designated enterprise based on the matching results of each identity type information.

[0131] In the examples of this specification, the above content can be implemented by the following specific implementation scheme:

[0132] Identity type information matching: For each identity type information, match the information of the specified enterprise with the corresponding information of multiple enterprises. The degree of matching can be determined using string matching, data comparison, or other suitable methods. The matching result of each identity type information is recorded, such as complete match, partial match, or no match.

[0133] Association index calculation: Assign a weight to each identity type information to reflect its importance in the association index. Calculate the association score of each enterprise with the designated enterprise on each identity type information based on the matching results and weights. Combine the association scores of each identity type information to calculate the total association index of each enterprise with the designated enterprise.

[0134] S108, identifying the enterprises whose association index is greater than a preset threshold as associated enterprises of the designated enterprise.

[0135] In the examples of this specification, the above content can be implemented by the following specific implementation scheme:

[0136] Determine the preset threshold: Determine an appropriate correlation index threshold based on specific business needs and correlation requirements. This threshold can be a fixed value or can be dynamically adjusted according to different situations.

[0137] Calculate the correlation index: Calculate the correlation index of multiple enterprises with the specified enterprise according to the method described above.

[0138] Compare relevance index to threshold: Compare the relevance index of each enterprise to the preset threshold.

[0139] Identify associated enterprises: If the association index of an enterprise is greater than a preset threshold, it is identified as an associated enterprise of the specified enterprise.

[0140] Result output: The list of identified related companies is output and can be used for further analysis, decision-making or other related purposes.

[0141] Furthermore, before matching the associated identification information with multiple identity type information in the associated identification information of multiple pre-screened enterprises, social network analysis technology can be used to analyze the preset relationship network between the designated enterprise and other enterprises, and screen out multiple enterprises associated with the designated enterprise. The preset relationship network includes investment relationships, cooperative relationships, and supply chain relationships between the designated enterprise and other enterprises.

[0142] It should be noted that the above content can be implemented through the following specific implementation plans:

[0143] Data collection: Collect relevant data of designated enterprises and other enterprises, including basic information of enterprises, investment relationships, cooperative relationships, supply chain relationships, etc. Data can be obtained from channels such as enterprise databases, financial institutions, industry reports, and news media.

[0144] Build a relationship network: Use social network analysis technology to build the collected data into a relationship network. In the relationship network, each enterprise is a node, and the relationship between enterprises (such as investment, cooperation, supply chain, etc.) is an edge.

[0145] Analyze the relationship network: Use social network analysis algorithms and tools to analyze the relationship network. You can calculate indicators such as the centrality, influence, and clustering coefficient of nodes to understand the position and role of the enterprise in the network. You can also identify key nodes and community structures in the network and find out the enterprise groups that are closely related to the specified enterprise.

[0146] Filter related companies: Based on the analysis results, filter out multiple companies that are related to the specified company. You can filter based on the following criteria:

[0147] An enterprise directly connected to a designated enterprise.

[0148] Firms with high centrality or influence in the network.

[0149] Businesses in the same community or cluster as the specified business.

[0150] Enterprises that have frequent investment, cooperation or supply chain relationships with designated enterprises.

[0151] Matching of associated identification information: Match the associated identification information of the selected associated enterprises with the associated identification information of multiple pre-screened enterprises. Match according to identity type information, such as enterprise name, unified social credit code, legal representative, etc. Data matching algorithms or tools can be used to improve the accuracy and efficiency of matching.

[0152] Result verification and optimization: Verify and evaluate the matching results to ensure that the selected related companies are accurate and reliable. Verification can be carried out through manual review, data comparison, etc. Based on the verification results, the implementation plan is optimized and adjusted to improve the accuracy and effect of the screening.

[0153] Through the above implementation plan, social network analysis technology can be used to analyze the preset relationship network between the specified enterprise and other enterprises, screen out multiple related enterprises, and match related identification information. This can help enterprises better understand their business ecosystem, discover potential partners, competitors or risk factors, and provide support for enterprise decision-making. The specific implementation details can be further refined and optimized according to actual conditions.

[0154] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0155] More comprehensive identification of related companies: Social network analysis technology can be used to discover relationships between companies that may be overlooked by traditional methods, such as indirect investments, cooperative alliances, etc., thereby more comprehensively identifying related companies.

[0156] Gain in-depth understanding of the enterprise ecosystem: By analyzing the preset relationship network, including investment relationships, partnerships, and supply chain relationships, we can gain a deeper understanding of the business ecosystem in which a given enterprise is located, as well as the interactions and dependencies between it and other enterprises.

[0157] Improve the accuracy of risk assessment: A comprehensive understanding of the relationships between enterprises helps to more accurately assess risks, as there may be risk transmission and mutual influence between related enterprises. This can help banks and related units better formulate risk management strategies.

[0158] Furthermore, before matching the associated identification information with multiple identity type information in the associated identification information of multiple pre-screened enterprises, a machine learning algorithm can be used to train and learn the associated identification information of the enterprise to obtain a neural network model of the screened enterprise associated with the enterprise; and multiple enterprises associated with the specified enterprise can be screened based on the neural network model.

[0159] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0160] Data preparation: Collect the associated identification information of the designated enterprise and other enterprises, including but not limited to the enterprise name, unified social credit code, legal representative, registered address, etc. Organize this information into a data set for training and learning the neural network model.

[0161] Data preprocessing: Clean and preprocess the data set, including noise removal, missing value processing, data standardization, etc. Divide the data set into training set, validation set and test set for model training, validation and evaluation.

[0162] Neural network model construction: Select a suitable neural network architecture, such as multi-layer perceptron (MLP), convolutional neural network (CNN) or recurrent neural network (RNN). According to the characteristics of the data set and task requirements, determine the number of neural network layers, number of neurons, activation function and other parameters. Use the training set to train the neural network, and continuously adjust the model parameters through the back propagation algorithm to improve the performance of the model.

[0163] Model evaluation and optimization: Use the validation set to evaluate the trained neural network model and calculate the model's accuracy, recall, F1 value and other indicators. Based on the evaluation results, optimize the model, such as adjusting the model's parameters, increasing the amount of training data, using a more complex neural network architecture, etc. Repeat the training and evaluation process until the model achieves satisfactory performance.

[0164] Related enterprise screening: Input the related identification information of the enterprise to be screened into the trained neural network model. The model outputs the enterprise prediction results related to the specified enterprise based on the learned patterns and features. Based on the prediction results, multiple enterprises related to the specified enterprise are screened out.

[0165] Result verification and adjustment: Manual verification of the selected related companies is carried out to ensure the accuracy of the results. If errors or inaccuracies are found in the screening results, the neural network model can be further optimized or adjusted in combination with other methods.

[0166] Continuous learning and updating: Over time, the relationships of an enterprise may change, so the dataset needs to be updated and the neural network model needs to be retrained regularly to maintain the accuracy and effectiveness of the model.

[0167] It should be noted that the embodiments of this specification have the following beneficial effects through the above contents:

[0168] Improved screening accuracy: Machine learning algorithms can automatically learn and identify patterns and features in enterprise-related identification information, thereby more accurately screening multiple enterprises associated with a given enterprise.

[0169] Adapting to complex relationships: The neural network model can handle complex enterprise relationship networks, including nonlinear relationships and multi-factor influences, and improve the ability to identify related enterprises.

[0170] Dynamically adapt to changes: By continuously updating training data, machine learning algorithms can adapt to the dynamic changes in corporate relationships and capture new associations in a timely manner.

[0171] It should be noted that corporate affiliation refers to the relationship between an enterprise and other enterprises that it directly or indirectly controls, as well as other relationships that may lead to the transfer of corporate interests. With the development of economic globalization and the intensification of market competition, the affiliation between enterprises has become more complex and hidden.

[0172] Due to the lack of identification of related companies by banks and related units, various risks and even actual losses are common. Therefore, how to collect data from enterprises and individual customers from multiple public sources in a legal and compliant manner to integrate, analyze and mine data to analyze the relationship between enterprises has become an important issue in today's financial risk prevention and control.

[0173] The purpose of the present invention is to provide a method and system for identifying corporate associations, which can accurately identify other companies that are suspected of being associated with the query company by comprehensively analyzing corporate business information, public annual report information, domain name information, patent information and software copyright information. The method and system are intended to enhance the risk prevention and control capabilities of banks and related units, reduce losses caused by insufficient identification of associated companies, and thus ensure the safety of financial credit assets.

[0174] The present invention provides a systematic method and tool to achieve the above objectives through the following steps:

[0175] 1. Data acquisition and preprocessing:

[0176] Obtain the business information, annual report information, domain name information, patent information and software copyright information of enterprises from public channels. Business annual report information refers to the annual report submitted by enterprises or individual industrial and commercial households to the industrial and commercial authorities through the market subject credit information disclosure system from January 1 to June 30 each year. These reports include basic information of enterprises, financial data, etc., and are disclosed to the public and can be queried by any unit or individual.

[0177] Clean and standardize the acquired data. Data cleaning includes deleting empty and dirty data of the enterprise, and standardization includes deleting spaces in the enterprise information and unifying the data format.

[0178] 2. Information dimension selection and cross comparison:

[0179] Select several enterprise information dimensions with public data and high quality for cross-comparison. These information dimensions include contact information, email address, mailing address, registered address, domain name, patent, software copyright, etc.

[0180] 3. Data matching and association identification:

[0181] Using the company's annual report data, search the annual report contact information table one by one for companies with the same contact information (contact information, email address, mailing address) in the current year or historical years, and regard them as suspected related companies.

[0182] Check the same patent information, find the patents applied by the main enterprise and check the information of other applicants. Enterprises that jointly apply for patents may be subsidiaries or joint ventures and are considered suspected related enterprises. Check the same software copyright information and identify suspected related enterprises based on similar methods.

[0183] The following is a detailed description of a specific embodiment of the present invention:

[0184] Obtain the business information and annual report information of enterprises from the market entity credit information disclosure system and other public channels, and store this information in a unified database.

[0185] Perform data cleaning, including deleting empty data and dirty data of the enterprise. Dirty data includes invalid data such as "None", "User-filled None", "0", "1", "0000000", "Flowing", etc. Standardize the data format and delete spaces and unnecessary symbols.

[0186] Use standardized contact information, email addresses, correspondence addresses, registered addresses and other information to cross-compare and query each in the annual report contact information table. Enterprises with the same contact information, email addresses, correspondence addresses and registered addresses in the current year or historical years are considered suspected affiliated enterprises.

[0187] Find the patents applied for by the main enterprise and check the information of other applicants. Analyze the enterprises that jointly apply for patents and identify possible subsidiaries or joint ventures as suspected related enterprises. Find the software copyright information registered by the main enterprise, cross-compare and query based on the same method, and identify suspected related enterprises.

[0188] Comprehensively analyze the analysis results of the above dimensions to further verify and confirm the authenticity of the suspected related companies. Sort all companies suspected of being related to the main company in descending order according to the number of hit information dimensions and divide them into three levels of association: highly suspected, moderately suspected, and slightly suspected. Companies that hit two or more suspected dimensions are classified as highly suspected, companies that only hit the same patents and the same software copyrights are classified as slightly suspected, and other companies are classified as moderately suspected. See Figure 2 A schematic diagram of association level determination is shown.

[0189] Output the final identification results of related enterprises, generate detailed reports, and provide them to banks and related units for risk assessment and decision support.

[0190] Improve risk prevention and control capabilities: Through multi-dimensional data analysis, related companies can be identified more accurately, thereby improving the risk prevention and control capabilities of banks and related units and reducing losses caused by insufficient identification of related companies.

[0191] High data processing efficiency: Through data cleaning and standardization, data quality is guaranteed, the efficiency of data matching and analysis is improved, and the need for manual intervention is reduced.

[0192] Legal and Compliance: Use publicly available corporate information to ensure the legality and compliance of methods and systems, and avoid privacy violations and violations of relevant regulations.

[0193] Multi-field application: This method and system are not only applicable to the financial field, but can also be used by regulatory agencies, market research institutions and other units that need to identify corporate relationships.

[0194] Figure 3 A schematic diagram of the structure of a device for identifying related enterprises provided for one or more embodiments of this specification includes: an identification information acquisition unit 302 , a matching unit 304 , a correlation index determination unit 306 and a related enterprise identification unit 308 .

[0195] The identification information acquisition unit 302 acquires the associated identification information of the specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information;

[0196] A matching unit 304 matches the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, and obtains matching results of each identity type information;

[0197] The association index determination unit 306 determines the association index between the plurality of enterprises and the designated enterprise based on the matching results of the identity type information;

[0198] The associated enterprise identification unit 308 identifies the enterprises whose association index is greater than a preset threshold as associated enterprises of the designated enterprise.

[0199] Furthermore, before matching the associated identification information with the plurality of identity type information in the associated identification information of the plurality of pre-screened enterprises, the device further comprises:

[0200] An interaction data acquisition unit, which acquires interaction data between the designated enterprise and other enterprises through multiple channels;

[0201] The interactive enterprise screening unit screens out a plurality of enterprises that interact with the designated enterprise from the interactive data.

[0202] Figure 4A schematic diagram of a structure of an associated enterprise identification device provided for one or more embodiments of this specification includes:

[0203] at least one processor; and,

[0204] a memory communicatively connected to the at least one processor; wherein,

[0205] The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0206] Obtaining the associated identification information of the specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information;

[0207] Matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, respectively, to obtain matching results of each identity type information;

[0208] Determining, based on the matching results of each identity type information, the association index between the multiple enterprises and the designated enterprise;

[0209] The enterprises whose association index is greater than a preset threshold are identified as associated enterprises of the designated enterprise.

[0210] One or more embodiments of this specification provide a non-volatile computer storage medium storing computer executable instructions, which can achieve the following when executed by a computer:

[0211] Obtaining the associated identification information of the specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information;

[0212] Matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, respectively, to obtain matching results of each identity type information;

[0213] Determining, based on the matching results of each identity type information, the association index between the multiple enterprises and the designated enterprise;

[0214] The enterprises whose association index is greater than a preset threshold are identified as associated enterprises of the designated enterprise.

[0215] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

Claims

1. A method for identifying an associated enterprise, characterized in that: The method comprises: Obtaining the associated identification information of the specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information; Matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, respectively, to obtain matching results of each identity type information; Determining, based on the matching results of each identity type information, the association index between the multiple enterprises and the designated enterprise; The enterprises whose association index is greater than a preset threshold are identified as associated enterprises of the designated enterprise.

2. The method according to claim 1, characterized in that Before matching the associated identification information with multiple identity type information in the associated identification information of multiple pre-screened enterprises, the method further includes: Obtaining interaction data between the designated enterprise and other enterprises through multiple channels; A plurality of enterprises that interact with the designated enterprise are screened out from the interaction data.

3. The method according to claim 1, characterized in that: The multiple identity type information includes one or more of business information, annual report information, domain name information and intellectual property information.

4. The method according to claim 1, characterized in that The associated identification information is matched with multiple identity type information in the associated identification information of multiple enterprises screened in advance, and matching results of each identity type information are obtained, including: Each piece of identity type information of the designated enterprise is compared with the corresponding identity type information of the multiple enterprises in turn to obtain the similarity between each piece of identity type information of the designated enterprise and the corresponding identity type information of each enterprise.

5. The method according to claim 1, characterized in that: Before matching the associated identification information with multiple identity type information in the associated identification information of multiple pre-screened enterprises, the method further includes: Social network analysis technology is used to analyze the preset relationship network between the designated enterprise and other enterprises, and multiple enterprises associated with the designated enterprise are screened out. The preset relationship network includes investment relationships, cooperation relationships, and supply chain relationships between the designated enterprise and other enterprises.

6. The method according to claim 1, characterized in that Before matching the associated identification information with multiple identity type information in the associated identification information of multiple pre-screened enterprises, the method further includes: Use machine learning algorithms to train and learn the associated identification information of enterprises, and obtain a neural network model for screening the associated enterprises of the enterprises; Based on the neural network model, multiple enterprises associated with the designated enterprise are obtained by screening.

7. A device for identifying an associated enterprise, characterized in that: include: An identification information acquisition unit, which acquires the associated identification information of a specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information; A matching unit, matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, to obtain matching results of each identity type information; an association index determination unit, which determines an association index between the plurality of enterprises and the designated enterprise based on a matching result of each identity type information; The associated enterprise identification unit identifies the enterprises whose association index is greater than a preset threshold as associated enterprises of the designated enterprise.

8. The device according to claim 7, characterized in that Before matching the associated identification information with the plurality of identity type information in the associated identification information of the plurality of enterprises screened in advance, the device further comprises: An interaction data acquisition unit, which acquires interaction data between the designated enterprise and other enterprises through multiple channels; The interactive enterprise screening unit screens out a plurality of enterprises that interact with the designated enterprise from the interactive data.

9. An associated enterprise identification device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Obtaining the associated identification information of the specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information; Matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, respectively, to obtain matching results of each identity type information; Determining, based on the matching results of each identity type information, the association index between the multiple enterprises and the designated enterprise; The enterprises whose association index is greater than a preset threshold are identified as associated enterprises of the designated enterprise.

10. A non-volatile computer storage medium, characterized in that: Computer executable instructions are stored, and when the computer executable instructions are executed by a computer, the following can be achieved: Obtaining the associated identification information of the specified enterprise, wherein the associated identification information is a plurality of pre-set identity type information; Matching the associated identification information with multiple identity type information in the associated identification information of multiple enterprises screened in advance, respectively, to obtain matching results of each identity type information; Determining, based on the matching results of each identity type information, the association index between the multiple enterprises and the designated enterprise; The enterprises whose association index is greater than a preset threshold are identified as associated enterprises of the designated enterprise.

Images