Small and micro enterprise credit data analysis method and system based on cloud computing and computer equipment

Abstract

The invention relates to the technical field of data analysis, in particular to a small and micro enterprise credit data analysis method and system based on cloud computing and computer equipment. According to the method, the access sensitivity threshold is compared with the information sensitivity matching table, the information with the sensitivity lower than or equal to the access sensitivity threshold is screened out and serves as the matching public information to be provided for the visitor, and the first credit score value is obtained through the first calculation model and the matching public information; the information of which the sensitivity is higher than the access sensitivity threshold value is used as hidden processing information, then the hidden processing information is processed in the node container to obtain a non-public score value, and the non-public score value is decrypted through the decryption key to obtain a second credit score value; and finally, obtaining a total credit score value according to the credit weighting coefficient, the first credit score value and the second credit score value, thereby realizing integration of credit scores of different sources, and enabling the total credit score to more objectively and accurately reflect the real credit level of small and micro enterprises.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a cloud computing-based small and micro enterprise credit data analysis method, system, and computer equipment. Background Art

[0002] Small and micro enterprises occupy an important position in the global economy and are an important driving force for economic growth. They play an irreplaceable role in promoting employment, promoting innovation, and increasing tax revenue. With the rapid development of information technology, small and micro enterprises generate a large amount of data in their daily operations. By analyzing the company's own credit data, a credit score can be obtained. The credit score can be used to assess the credit status of small and micro enterprises. The credit score can not only enable small and micro enterprises to have an in-depth understanding of the advantages and disadvantages of their own operating conditions and credit status, but also allow financial institutions to intuitively understand the repayment ability and default risk of small and micro enterprises, which helps them to reasonably allocate credit resources. At the same time, it can also encourage local governments to implement appropriate support policies for small and micro enterprises.

[0003] In the current process of analyzing the credit data of small and micro enterprises and obtaining credit scores, visitors with different identities have different access rights when accessing the credit data of small and micro enterprises. Therefore, the content of the credit data obtained is different. This difference in the content obtained makes the final credit score results unable to accurately and objectively reflect the credit status of small and micro enterprises. Summary of the Invention

[0004] The main purpose of the present invention is to provide a small and micro enterprise credit data analysis method, system and computer equipment based on cloud computing, aiming to solve the technical problems in the prior art.

[0005] The present invention proposes a small and micro enterprise credit data analysis method based on cloud computing, comprising:

[0006] Obtaining the visitor's identity information, and obtaining the visit purpose information and access permission information based on the identity information;

[0007] Acquiring evaluation core information according to the access purpose information, wherein the evaluation core information includes evaluation model information and multiple core data information, and acquiring an information sensitivity matching table according to the multiple core data information;

[0008] Obtaining the access sensitivity threshold of the visitor according to the access permission information, and screening multiple core data information according to the access sensitivity threshold and the information sensitivity matching table to obtain corresponding multiple matching public information and multiple hidden processing information;

[0009] Obtaining a first calculation model based on the evaluation model information and a plurality of matching public information, and obtaining a first credit score value based on the first calculation model and the plurality of matching public information;

[0010] Transmitting multiple hidden processing information to the node container for calculation according to the evaluation model information to obtain a hidden calculation value, and encrypting the hidden calculation value according to the access purpose information to obtain a non-public rating value;

[0011] Obtaining a decryption key based on the identity information, and obtaining a second credit score value based on the decryption key and the non-public score value;

[0012] A credit weighting coefficient is obtained according to the access sensitivity threshold, and a total credit score value is obtained according to the credit weighting coefficient, the first credit score value, and the second credit score value.

[0013] Preferably, the step of obtaining an information sensitivity matching table according to the plurality of core data information includes:

[0014] Acquire corresponding historical collection information and information update cycle according to the plurality of core data information, wherein the historical collection information includes historical access frequency, visitor type frequency, and business process association frequency;

[0015] Obtaining a collection frequency value based on the historical access frequency, visitor type frequency, and business process association frequency, and scoring multiple core data information based on the collection frequency value to obtain a corresponding confidentiality score value;

[0016] Scoring multiple core data information according to the information update cycle to obtain a timeliness score value;

[0017] The sensitivity values ​​corresponding to the multiple core data information are obtained according to the confidentiality score value and the timeliness score value, and the multiple core data information are classified according to the sensitivity values ​​to obtain an information sensitivity matching table.

[0018] Preferably, the step of obtaining the visitor's access sensitivity threshold according to the access permission information includes:

[0019] Acquiring operation authority information according to the access authority information, wherein the operation authority information includes information reading, information editing, and information approval;

[0020] Determine each operation permission according to the operation permission information, obtain operation result information, and obtain a corresponding sensitivity threshold interval according to the operation result information;

[0021] Acquiring operation detail information according to the operation authority information, wherein the operation detail information includes operation duration and operation location, and acquiring a first weight coefficient according to the operation duration;

[0022] Obtaining a location value according to the operation location, and obtaining a second weight coefficient according to the location value;

[0023] An access sensitivity threshold is obtained according to the sensitivity threshold interval, the operation duration, the first weight coefficient, the location assignment, and the second weight coefficient.

[0024] Preferably, the step of obtaining a first calculation model based on the evaluation model information and a plurality of matching public information, and obtaining a first credit score value based on the first calculation model and the plurality of matching public information includes:

[0025] Acquire evaluation dimension information and evaluation logic information according to the evaluation model information, and acquire multiple key factor information according to the evaluation dimension information and multiple matching public information;

[0026] Obtaining key sensitivity values ​​corresponding to the plurality of key factor information, and prioritizing the plurality of key factor information according to the key sensitivity values ​​to obtain key factor ranking information, and obtaining indicator weight parameters corresponding to the key factor information according to the evaluation logic information and the key factor ranking information;

[0027] Constructing a first calculation model according to the evaluation logic information and indicator weight parameters;

[0028] Corresponding key factor values ​​are obtained according to the plurality of key factor information, and a first credit score value is obtained according to the plurality of key factor values ​​and a first calculation model.

[0029] Preferably, the step of transmitting the plurality of hidden processing information to the node container for calculation according to the evaluation model information to obtain the hidden calculation value includes:

[0030] Acquire a second calculation model according to the evaluation model information, and acquire the number of scoring factor input layers and the number of scoring associated processing layers according to the second calculation model;

[0031] Acquire a first bias value and a second bias value according to the second calculation model;

[0032] Acquire multiple scoring factor values ​​according to the anonymized information, and acquire corresponding first connection weights according to the scoring factor values;

[0033] Obtaining a scoring factor summary value according to the scoring factor value, the first connection weight, and the first bias value;

[0034] Performing a linear transformation on the summary value of the scoring factors to obtain an input value of the scoring association processing layer;

[0035] A corresponding second connection weight is obtained according to the score-associated processing layer input value, and a hidden calculation value is obtained according to the score-associated processing layer input value, the second connection weight and the second bias value.

[0036] Preferably, the step of encrypting the hidden calculated value according to the access purpose information to obtain a non-public rating value includes:

[0037] Acquiring business scenario information according to the access purpose information, and acquiring encryption requirement information according to the business scenario information, wherein the encryption requirement information includes security level requirements, resource limitation requirements, and processing efficiency requirements;

[0038] Acquire an encryption key length that matches the security level requirement, and acquire an adaptive encryption mode based on the resource limitation requirement;

[0039] Obtaining an adapted encryption algorithm according to the processing efficiency requirement;

[0040] The multiple hidden calculation values ​​are encrypted according to the encryption key length, encryption mode and encryption algorithm to obtain multiple non-public scoring values.

[0041] This application also provides a small and micro enterprise credit data analysis system based on cloud computing, including:

[0042] A first acquisition module is used to obtain the identity information of the visitor, and obtain the visit purpose information and access permission information based on the identity information;

[0043] a second acquisition module, configured to acquire evaluation core information according to the access purpose information, wherein the evaluation core information includes evaluation model information and a plurality of core data information, and acquire an information sensitivity matching table according to the plurality of core data information;

[0044] a screening module, configured to obtain an access sensitivity threshold of the visitor based on the access permission information, and screen multiple core data information based on the access sensitivity threshold and the information sensitivity matching table to obtain corresponding multiple matching public information and multiple hidden processing information;

[0045] a first calculation module, configured to obtain a first calculation model based on the evaluation model information and a plurality of matching public information, and obtain a first credit score value based on the first calculation model and the plurality of matching public information;

[0046] A second calculation module is configured to transmit the plurality of hidden processing information to the node container for calculation according to the evaluation model information to obtain a hidden calculation value, and encrypt the hidden calculation value according to the access purpose information to obtain a non-public rating value;

[0047] a decryption module, configured to obtain a decryption key based on the identity information, and obtain a second credit score value based on the decryption key and the non-public score value;

[0048] The third calculation module is configured to obtain a credit weighting coefficient according to the access sensitivity threshold, and obtain a total credit score value according to the credit weighting coefficient, the first credit score value, and the second credit score value.

[0049] Preferably, the second acquisition module includes:

[0050] A first acquiring unit is configured to acquire corresponding historical collection information and information update period according to the plurality of core data information, wherein the historical collection information includes historical access frequency, visitor type frequency, and business process association frequency;

[0051] A first scoring unit is configured to obtain a collection frequency value based on the historical access frequency, visitor type frequency, and business process association frequency, and score multiple core data information based on the collection frequency value to obtain a corresponding confidentiality score value;

[0052] A second scoring unit is used to score the multiple core data information according to the information update cycle to obtain a timeliness score value;

[0053] The classification unit is used to obtain sensitivity values ​​corresponding to multiple core data information according to the confidentiality score value and the timeliness score value, and classify the multiple core data information according to the sensitivity values ​​to obtain an information sensitivity matching table.

[0054] The present invention also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned cloud computing-based small and micro enterprise credit data analysis method are implemented.

[0055] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned cloud computing-based small and micro enterprise credit data analysis method.

[0056] The beneficial effects of the present invention are as follows: the present invention obtains the visitor's access purpose information and access permission information based on the visitor's identity information. The access purpose information determines what information the visitor needs to obtain and the types of operations that may be performed. The access permission information stipulates which resources the visitor can access and the permission information for what operations can be performed on these resources. Then, the required evaluation core information is obtained according to the access purpose. The evaluation core information includes evaluation model information for calculating the credit score value and core data information for inputting the evaluation model. The obtained core data information is evaluated for sensitivity value, and the core data information is classified according to the sensitivity value and an information sensitivity matching table is generated. It clarifies the sensitivity of core data information with different sensitivity values, so that the system can provide information that meets the visitor's access permission according to the visitor's access sensitivity threshold. Then, the access sensitivity threshold is compared with the information sensitivity matching table to screen out information with a sensitivity lower than or equal to the access sensitivity threshold and provide it to the visitor as matching public information, while information with a sensitivity higher than the access sensitivity threshold is treated as hidden processing information. Through this information screening method, fine permission management of visitors is achieved to ensure that only visitors with corresponding permissions can obtain specific sensitive information. The first calculation model is constructed by using the evaluation model information and the matching public information. The matching public information is analyzed and calculated by the first calculation model to obtain a first credit score value. The first credit score value reflects the credit status of small and micro enterprises based on public data. The hidden processing information is then transmitted to the node container, and the distributed computing capability of cloud computing is used to perform secure computing on the sensitive data to obtain a hidden calculation value. The encryption strategy is then customized according to the access purpose information, and the hidden calculation value is encrypted to generate a non-public score value, thereby preventing the sensitive information from being stolen or tampered with during transmission and storage. The identity information of the questioner is used to obtain the corresponding decryption key, and the decryption key is used to decrypt the non-public score value to obtain a second credit score value. The second credit score value reflects the credit status of small and micro enterprises based on sensitive data. Finally, the credit weighting coefficient is determined according to the access sensitivity threshold. The credit weighting coefficient reflects the relative importance of matching public information and concealed information in the total credit score. Finally, the total credit score value is obtained according to the credit weighting coefficient, the first credit score value and the second credit score value, which realizes the reasonable integration of credit scores from different sources, so that the total credit score more objectively and accurately reflects the true credit level of small and micro enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 Schematic diagram of a method flow according to an embodiment of the present invention.

[0058] Figure 2 FIG. 1 is a schematic diagram of a system structure according to an embodiment of the present invention.

[0059] Figure 3 This is a schematic diagram of the internal structure of a computer device according to an embodiment of the present application.

[0060] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0061] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] like Figure 1 As shown, this application provides a small and micro enterprise credit data analysis method based on cloud computing, including:

[0063] S1. Obtain the visitor's identity information, and obtain the visit purpose information and access permission information based on the identity information;

[0064] S2. Acquire evaluation core information according to the access purpose information, wherein the evaluation core information includes evaluation model information and multiple core data information, and acquire an information sensitivity matching table according to the multiple core data information;

[0065] S3. Obtaining an access sensitivity threshold of the visitor based on the access permission information, and screening multiple core data information based on the access sensitivity threshold and the information sensitivity matching table to obtain corresponding multiple matching public information and multiple hidden processing information;

[0066] S4. Obtaining a first calculation model based on the evaluation model information and multiple matching public information, and obtaining a first credit score value based on the first calculation model and multiple matching public information;

[0067] S5. Transmitting the plurality of hidden processing information to the node container for calculation according to the evaluation model information to obtain a hidden calculation value, and encrypting the hidden calculation value according to the access purpose information to obtain a non-public rating value;

[0068] S6. Obtain a decryption key based on the identity information, and obtain a second credit score value based on the decryption key and the non-public score value;

[0069] S7. Obtain a credit weighting coefficient according to the access sensitivity threshold, and obtain a total credit score value according to the credit weighting coefficient, the first credit score value, and the second credit score value.

[0070] As described in the above steps S1-S7, small and micro enterprises occupy an important position in the global economy and are an important driving force for economic growth. They play an irreplaceable role in promoting employment, promoting innovation, and increasing tax revenue. With the rapid development of information technology, small and micro enterprises generate a large amount of data in their daily operations. By analyzing the company's own credit data, a credit score value can be obtained. The credit score value can be used to evaluate the credit status of small and micro enterprises. The credit score value can not only enable small and micro enterprises to have an in-depth understanding of the advantages and disadvantages of their own operating conditions and credit status, but also allow financial institutions to intuitively understand the repayment ability and default risk of small and micro enterprises, which helps them to reasonably allocate credit resources. At the same time, it can also encourage local governments to implement appropriate support policies for small and micro enterprises. In the current process of analyzing the credit data of small and micro enterprises and obtaining credit score values, visitors with different identities have different access rights when accessing the credit data of small and micro enterprises due to the differences in their identities. Therefore, the content of the credit data obtained is different. This difference in the content obtained makes the final credit score result unable to accurately and objectively reflect the credit situation of small and micro enterprises. The present invention obtains the identity information of the visitor to clarify the identity of the visitor, laying the foundation for subsequent access control and information provision. Among them, the identity information refers to the relevant data used to identify the identity of the visitor, including name, ID number, company name, unified social credit code, etc. Then, the visitor's access purpose information and access right information are obtained. Among them, the access purpose information refers to the visitor's intention to perform the access operation, such as querying the company's financial data, obtaining business contract information, applying for loan amount assessment, etc. The access purpose information determines what information the visitor needs to obtain and the type of operation that may be performed. It is an important basis for determining the processing method in the subsequent process. The access right information refers to the information that specifies the visitor's access rights. The permission information on which resources can be accessed and what operations can be performed on these resources is obtained, and then the required core evaluation information is obtained according to the purpose of access. The core evaluation information refers to the collection of data information and model information that plays a key role in the credit evaluation process of small and micro enterprises, including evaluation model information and core data information. The evaluation model information refers to the collection of calculation models used to calculate credit score values ​​for core data information with different sensitivity values, as well as the various parameters, rules and structures contained in the calculation model construction. The evaluation model information defines how to process and calculate the input core data information to obtain the final credit evaluation result. The core data information refers to all basic data that are closely related to the credit evaluation of small and micro enterprises and constitute the important input data of the evaluation model. The obtained core data information is then evaluated for sensitivity value, and the information is classified according to the sensitivity value (such as low sensitivity value information, medium sensitivity value information, and high sensitivity value information) to generate an information sensitivity matching table. The information sensitivity matching table is a table that associates and matches different information with corresponding sensitivity value levels.It clarifies the sensitivity of various core assessment information, so that the system can provide appropriate information according to the visitor's permissions and sensitivity thresholds, and then determines the sensitivity value range of the information that the visitor can access, and obtains the access sensitivity threshold, where the access sensitivity threshold refers to a sensitivity limit value set for each visitor or access role. Then, the access sensitivity threshold is compared with the information sensitivity matching table, and information with a sensitivity value lower than or equal to the access sensitivity threshold is screened out and provided to the visitor as matching public information, while information with a sensitivity value higher than the access sensitivity threshold is treated as hidden information, where matching public information refers to data that can be disclosed to the visitor that is screened out from the core data information, and hidden information is processed. Refers to the data that cannot be disclosed to the visitor that is screened out from the core data information. This information screening method realizes the fine permission management of the visitor, ensuring that only visitors with corresponding permissions can obtain information of specific sensitivity. Then, the first calculation model is constructed using the evaluation model information and the matching public information. The first calculation model refers to the calculation model for processing the matching public information that is screened out from the evaluation model information. The matching public information is analyzed and calculated by the first calculation model to obtain the first credit score value, which reflects the credit status of small and micro enterprises based on the public data. Then, the hidden processing information is transmitted to the node container, and the distributed computing power of cloud computing is used to perform secure computing on the sensitive data to obtain Anonymous calculation value ensures the security of sensitive data during the calculation process. The anonymous calculation value refers to the value obtained by transferring information that is not suitable for direct disclosure but is indispensable to the node container for calculation. Then, the encryption strategy is customized according to the access purpose information to improve the pertinence and effectiveness of the encryption. The anonymous calculation value is encrypted through the customized encryption strategy to generate a non-public scoring value to prevent sensitive information from being stolen or tampered with during transmission and storage. The non-public scoring value refers to the value obtained after encrypting the anonymous calculation value. Then, the corresponding decryption key is obtained based on the visitor's identity information. Only legitimate visitors whose identity information has been verified can obtain the correct decryption key to ensure the decryption operation The legitimacy and security of the operation are ensured, and then the non-public score value is decrypted using the decryption key to obtain a second credit score value, where the second credit score value refers to the credit score value obtained after decrypting the non-public score value using the decryption key. Through strict decryption authority management, the security of sensitive data is guaranteed. Only authorized visitors can obtain the credit score corresponding to the sensitive data, and the second credit score value reflects the credit status of small and micro enterprises based on sensitive data. Finally, the credit weighting coefficient is determined based on the access sensitivity threshold, where the credit weighting coefficient refers to the coefficient used to measure the relative importance of the first credit score value and the second credit score value in calculating the total credit score value, including the first weight coefficient and the second weight coefficient.The first and second weighting coefficients reflect the relative importance of matching public information and concealed information in the overall credit score, respectively. The lower the access sensitivity threshold (lower the authority), the lower the first weighting coefficient corresponding to matching public information, and the higher the second weighting coefficient corresponding to concealed information. The formula for calculating the first weighting coefficient is: first weighting coefficient = access sensitivity threshold ÷ 100, while the formula for calculating the second weighting coefficient is: second weighting coefficient = 1 - first weighting coefficient. The first and second weighting coefficients are then used to weight the first and second credit scores to obtain the total credit score. This achieves a reasonable integration of credit scores from different sources, making the total credit score more objective and accurate in reflecting the true credit level of small and micro enterprises.

[0071] In one embodiment, the step S2 of obtaining an information sensitivity matching table according to the plurality of core data information includes:

[0072] S21. Acquire corresponding historical collection information and information update cycle according to the plurality of core data information, wherein the historical collection information includes historical access frequency, visitor type frequency, and business process association frequency;

[0073] S22. Obtaining a collection frequency value based on the historical access frequency, visitor type frequency, and business process association frequency, and scoring multiple core data information based on the collection frequency value to obtain corresponding confidentiality score values;

[0074] S23. Scoring the multiple core data information according to the information update cycle to obtain a timeliness score;

[0075] S24. Obtain sensitivity values ​​corresponding to multiple core data information according to the confidentiality score value and the timeliness score value, and classify the multiple core data information according to the sensitivity values ​​to obtain an information sensitivity matching table.

[0076] As described in the above steps S21-S24, the present invention obtains the historical collection information and information update cycle corresponding to the core data information, wherein the historical collection information refers to a series of related data generated and recorded in the past collection process of various core data information, which includes historical access frequency, visitor type frequency and business process association frequency. The historical access frequency refers to the frequency of access to specific core data information in the past period of time, the visitor type frequency refers to the frequency of access to specific core data information by different types of visitors, the business process association frequency refers to the closeness and frequency of association between a certain core data information and various business processes of the enterprise, and the information update cycle Period refers to the time interval between the completion of one update and the completion of the next update of the core data information. These data can reflect the value of the core data information from different angles. Then, the historical access frequency, visitor type frequency and business process association frequency are standardized respectively. The standardized processing formula for the historical access frequency of specific core data information is: Standardized historical access frequency of specific core data information = (historical access frequency of specific core data information - minimum value of historical access frequency) / (maximum value of historical access frequency - minimum value of historical access frequency). Similarly, the standardized visitor type frequency of specific core data information and the standardized business process association frequency of specific core data information are obtained respectively. , then further calculate the collection frequency value corresponding to the specific core data information based on the acquired data, where the calculation formula is: the collection frequency value corresponding to the specific core data information = the standardized historical access frequency of the specific core data information * the first weight + the standardized visitor type frequency of the specific core data information * the second weight + the standardized business process association frequency of the specific core data information * the third weight, and then map the collection frequency value to the scoring interval to obtain the information confidentiality score value, for example: 0 ≦ collection frequency value ﹤ 0.2, in terms of information confidentiality, it can be scored 20 points, 0.2 ≦ collection frequency value ﹤ 0.4, in terms of information confidentiality, it can be scored 40 points, 0.4 ≦ collection frequency value ﹤ 0.6, in The information confidentiality level can be scored as 60 points, 0.6 ≦ collection frequency value < 0.8, the information confidentiality level can be scored as 80 points, 0.8 ≦ collection frequency value < 1, the information confidentiality level can be scored as 100 points. Through the scoring mechanism, each core data information has a clear quantitative value of confidentiality level. A high score indicates that the information is highly confidential and requires more stringent protection measures. A low score indicates that the information is relatively less confidential and a relatively loose management method can be adopted. Then, multiple core data information are scored according to the information update cycle. For example, assuming that the update cycle ranges from 1 day to 30 days, we map this range to the scoring interval of 0-100 points, and the scoring equation is: When the update cycle is 1 day, the timeliness score is 100, and when the update cycle is 30 days, the timeliness score is 0. Information with high timeliness may have higher sensitivity due to the timeliness and importance of its content, which helps to distinguish the differences in timeliness of different information. Finally, the sensitivity value corresponding to each core data information is calculated based on the confidentiality score and the timeliness score. The calculation formula is: sensitivity value = information confidentiality score * 0.5 + timeliness score * 0.5. Finally, the core data information is classified according to the sensitivity value, and the classified information is organized into an information sensitivity matching table to provide an intuitive reference basis for subsequent information security management and data usage decisions.

[0077] In one embodiment, the step S3 of obtaining the access sensitivity threshold of the visitor according to the access permission information includes:

[0078] S31. Acquire operation permission information according to the access permission information, wherein the operation permission information includes information reading, information editing, and information approval;

[0079] S32: Determine each operation permission according to the operation permission information, obtain operation result information, and obtain a corresponding sensitivity threshold interval according to the operation result information;

[0080] S33. Acquire operation detail information according to the operation authority information, wherein the operation detail information includes operation duration and operation location, and acquire a first weight coefficient according to the operation duration;

[0081] S34. Obtaining a location value according to the operation location, and obtaining a second weight coefficient according to the location value;

[0082] S35. Obtain an access sensitivity threshold according to the sensitivity threshold interval, the operation duration, the first weight coefficient, the location assignment, and the second weight coefficient.

[0083] As described in steps S31-S35 above, the present invention obtains the visitor's operation permission information based on the access permission information, wherein the operation permission information refers to the specific permission content related to the operations that the visitor can perform in the system or specific environment, including information reading, information editing and information approval permissions. Then, each operation permission is judged separately to determine whether the corresponding operation can be performed. If it can be performed, its result is marked as 1, and if it cannot be performed, it is marked as 0. Through this judgment method, a series of operation result information will be obtained, such as (1, 1, 1), (1, 1, 0), (1, 0, 0), (0, 0, 0), etc., wherein the result is (1, 1, 1). The result of (1, 1, 0) means that the access right has complete information reading, editing and approval capabilities, and the corresponding access sensitivity threshold is in the highest range, which is [85, 100]. The result of (1, 1, 0) means that the access right has read and edit permissions but no approval permissions, and the corresponding access sensitivity threshold range is (75, 90). The result of (0, 0, 0) means that there is no permission for any of the above operations, and the corresponding access sensitivity threshold is in the lowest range, which is [45, 60]. According to the specific operation permissions of the visitor, the sensitivity range corresponding to the information that the visitor can access can be preliminarily determined, and then the operation details information is obtained, among which the operation details information Refers to other detailed information about the visitor's execution of relevant operation permissions, such as operation duration and operation location, and then quantifies and grades the operation duration. For example, the operation duration is divided into short time (less than 1 hour), with a score of 10-30, medium time (1-8 hours), with a score of 30-60, and long time (more than 8 hours), with a score of 60-100. The specific score value is obtained according to the specific operation duration, and then the operation location is assigned a location value, which is divided into specific areas within the company, all areas within the company, specific security areas outside the company, and unrestricted areas outside the company. The corresponding assigned scores are 20, 40, 60, and 100 respectively. The operation duration and operation location are divided into specific areas within the company, all areas within the company, specific security areas outside the company, and unrestricted areas outside the company. The corresponding assigned scores are 20, 40, 60, and 100 respectively. Detailed information such as the operation location can reflect the depth or complexity of the visitor's information processing. Finally, the access sensitivity threshold is calculated based on the sensitivity threshold interval, operation time, the first weight coefficient, the location assignment and the second weight coefficient. The calculation formula is: Sensitivity threshold = lowest value of the sensitivity threshold interval + operation time * first weight coefficient + location assignment * second weight coefficient. The sensitivity threshold obtained by calculation can determine the sensitivity range corresponding to the relevant information that the visitor can access, thereby preventing visitors with low access levels from accessing the company's confidential core data information, which is conducive to ensuring that the company's confidential information is only disclosed to specific personnel and preventing the leakage of the company's confidential information.

[0084] In one embodiment, the step S4 of obtaining a first calculation model based on the evaluation model information and a plurality of matching public information, and obtaining a first credit score value based on the first calculation model and the plurality of matching public information, includes:

[0085] S41. Acquire evaluation dimension information and evaluation logic information according to the evaluation model information, and acquire multiple key factor information according to the evaluation dimension information and multiple matching public information;

[0086] S42: Acquire key sensitivity values ​​corresponding to the plurality of key factor information, prioritize the plurality of key factor information according to the key sensitivity values ​​to obtain key factor ranking information, and acquire indicator weight parameters corresponding to the key factor information according to the evaluation logic information and the key factor ranking information;

[0087] S43: Construct a first calculation model based on the evaluation logic information and indicator weight parameters;

[0088] S44. Obtain corresponding key factor values ​​according to the plurality of key factor information, and obtain a first credit score value according to the plurality of key factor values ​​and a first calculation model.

[0089] As described in the above steps S41-S44, the present invention extracts evaluation dimension information and evaluation logic information from the evaluation model information, wherein the evaluation dimension information refers to the relevant content of various aspects or angles based on the evaluation process, which is used to clarify which aspects of small and micro enterprises (such as financial, operational, market and other dimensions) to evaluate credit, and the evaluation logic information refers to information on rules, methods and ideas for analyzing, judging and drawing conclusions on the evaluation objects, which stipulates the relationship between various evaluation factors and how to calculate or infer the final evaluation results based on these factors. Based on the evaluation dimension information, multiple matching public information are screened and refined to find out the key factors that can effectively reflect the credit status of small and micro enterprises. Sub-information, where key factor information refers to the relevant data corresponding to factors that have a significant impact and decisive role on the evaluation results, such as debt-to-asset ratio, operating data, default status, etc., closely screening the matching public information around the dimensions set by the evaluation model can ensure that the selection of key factor information is targeted, so that the subsequent credit score calculation is more in line with the evaluation target. For example, if the evaluation model is a linear weighted model, then the key factor information is the key factor information such as market share and customer complaint rate in the matching public information. Since different key factors have different importance to the credit evaluation of small and micro enterprises, they are prioritized by sensitivity value to obtain key factor ranking information, where the key factor ranking information is Refers to information that arranges key factors according to their sensitivity. The key factor ranking information can intuitively present the relative importance of each factor. Then, based on the evaluation logic information, the weight of each key factor in the credit assessment is determined, thereby obtaining the indicator weight parameters. For example, when the evaluation logic emphasizes financial soundness, key financial factors such as the debt-to-asset ratio are given a higher weight to achieve differentiated treatment of key factors, highlight the role of important factors in credit scoring, and make the credit scoring results more accurately reflect the actual credit level of the enterprise. Then, the evaluation logic information and the determined indicator weight parameters are combined to construct the first calculation model suitable for matching public information. The first calculation model clarifies how to use key factors. The factor information is combined according to specific weights and calculation rules to obtain a credit score. Finally, the key factor value is obtained based on the multiple key factor information screened out. The key factor value refers to the specific numerical value or data performance corresponding to the key factor. It is a quantitative reflection of the key factor on the actual evaluation object. The specific numerical value is used to reflect the state or degree of the key factor in a specific situation. For example, if the debt-to-asset ratio is used as a key factor, then the specific calculation result of the debt-to-asset ratio, the debt-to-asset ratio index value, is used as the key factor value. These key factor values ​​are input into the constructed first calculation model, and the first credit score value is obtained according to the calculation rules and weight distribution set by the model. For example, the function expression of the first calculation model can be Where S1 refers to the first credit score value, n refers to the number of key factor values, and x i refers to the value of the i-th key factor, ω i It refers to the indicator weight parameter corresponding to the i-th key factor value, where i represents the serial number of the key factor value. The first credit score value can be calculated by the first calculation model. In addition, the first calculation model can also be a linear weighted model or other evaluation model that is more suitable for the credit evaluation of small and micro enterprises after adjusting certain parameters, such as the Z-score model, Logistic regression model, machine learning model, etc. The evaluation model is not uniquely limited here. The first credit score value reflects the credit status of small and micro enterprises based on matching public information, realizes the conversion from data information to specific credit scores, and provides a quantitative preliminary result for the credit evaluation of small and micro enterprises, which is convenient for intuitively understanding the credit level of enterprises based on public information.

[0090] In one embodiment, the step S51 of transmitting the plurality of hidden processing information to the node container for calculation according to the evaluation model information to obtain the hidden calculation value includes:

[0091] S511. Acquire a second calculation model according to the evaluation model information, and acquire the number of scoring factor input layers and the number of scoring-related processing layers according to the second calculation model;

[0092] S512. Acquire a first bias value and a second bias value according to the second calculation model;

[0093] S513: Acquire multiple scoring factor values ​​according to the anonymized information, and acquire corresponding first connection weights according to the scoring factor values;

[0094] S514: Obtain a scoring factor summary value based on the scoring factor value, the first connection weight, and the first bias value, wherein the calculation formula is:

[0095]

[0096] Among them, h j represents the summary value of the scoring elements of the j-th scoring association processing layer, n represents the number of scoring element input layers, ω ij represents the first connection weight from the i-th scoring factor input layer to the j-th scoring association processing layer, I i Indicates the scoring factor value corresponding to the i-th scoring factor, b i represents the first bias value corresponding to the i-th scoring factor input layer, i represents the serial number of the scoring factor value, and j represents the serial number of the scoring associated processing layer;

[0097] S515: Perform a linear transformation on the summary value of the scoring factors to obtain the input value of the scoring association processing layer, wherein the calculation formula is:

[0098]

[0099] Among them, H j represents the input value of the j-th score association processing layer, h j represents the summary value of the scoring elements of the j-th scoring associated processing layer, and j represents the sequence number of the scoring associated processing layer;

[0100] S516. Obtain a corresponding second connection weight according to the score-associated processing layer input value, and obtain a hidden calculation value according to the score-associated processing layer input value, the second connection weight, and the second bias value, wherein the calculation formula is:

[0101]

[0102] Among them, S represents the hidden calculation value, m represents the number of score-related processing layers, ω j represents the second connection weight from the jth score association processing layer to the output layer, H j represents the input value of the j-th score association processing layer, b j represents the second bias value corresponding to the j-th score-associated processing layer, and j represents the sequence number of the score-associated processing layer.

[0103] As in the above steps S511-S516, the present invention obtains a second calculation model according to the evaluation model information, wherein the second calculation model refers to a model obtained from the evaluation model information for calculating multiple hidden processing information, such as a neural network model. According to the second calculation model, the number of scoring factor input layers and the number of scoring associated processing layers are clarified, thereby building a framework for subsequent calculations, determining the dimensions of data input in the model and the number of intermediate processing layers, so that the calculation process has a clear structure and direction, wherein the scoring factor input layer refers to the part of the second calculation model that receives input data. The scoring factor input layer is like the "entrance" of the model, which is responsible for introducing the original scoring-related data into the calculation system of the model for subsequent processing and analysis. Provide data source, the scoring association processing layer is located after the scoring factor input layer, and is an intermediate layer for further processing and processing the input scoring factor values. In the scoring association processing layer, the input data is operated through specific calculation logic (such as weighted summation) to extract and mine the features and relationships related to the score in the data. Clarifying the number of input layers and processing layers helps to reasonably match the input data with the model structure, ensure that the data can flow and process in the model in a predetermined manner, and lay the foundation for accurate calculation. The number of input layers and processing layers can be flexibly adjusted according to different evaluation requirements and data characteristics, so that the model can better adapt to various complex situations and improve the versatility and adaptability of the model. Then, according to the second calculation The model obtains the first bias value and the second bias value. The first bias value and the second bias value are an adjustable constant term added to the calculation of the scoring factor input layer and the scoring association processing layer respectively. By introducing the bias value, the model can express a more complex functional relationship, not just limited to the linear combination of input data. It can affect the output results of the scoring factor input layer and the scoring association processing layer without relying on the input data, so that the model has stronger fitting and generalization capabilities. Then, multiple scoring factor values ​​are obtained according to the hidden processing information, where the scoring factor value refers to the value of each specific factor related to the scoring of the evaluation object, such as profit margin, etc. Then, the first connection weight corresponding to the scoring factor value is obtained. The first connection weight determines each The degree of influence of the scoring factor on subsequent calculations can highlight the importance difference of different scoring factors in the model, so that the model can pay more attention to the factors that have a greater impact on the results, and obtain the scoring factor summary value based on the scoring factor value, the first connection weight and the first bias value. The scoring factor summary value refers to the result obtained by integrating the information of multiple scoring factors and preliminarily integrating and processing the input scoring factors. This value comprehensively considers the influence of all scoring factors. Then, the scoring factor summary value is linearly transformed to obtain the input value of the scoring association processing layer. The linear transformation can map the summary value to a suitable range or space, so that the data is more in line with the requirements of subsequent model processing. Then, the corresponding second connection weight is obtained according to the input value of the scoring association processing layer.The second connection weight refers to the connection parameter between the input value of the scoring association processing layer and the calculated hidden calculation value. Based on the input value of the scoring association processing layer, it further determines the importance of these output values ​​in the final calculation of the hidden calculation value. Finally, the hidden calculation value is obtained based on the input value of the scoring association processing layer, the second connection weight, and the second bias value. The obtained hidden calculation value can serve as an important basis for credit assessment of small and micro enterprises based on the hidden processing information, providing strong support for subsequent analysis.

[0104] In one embodiment, the step S52 of encrypting the hidden calculated value according to the access purpose information to obtain the non-public rating value includes:

[0105] S521. Acquire business scenario information according to the access purpose information, and acquire encryption requirement information according to the business scenario information, wherein the encryption requirement information includes security level requirements, resource limitation requirements, and processing efficiency requirements;

[0106] S522. Acquire an encryption key length that matches the security level requirement, and acquire an adaptive encryption mode according to the resource limitation requirement;

[0107] S523. Obtain an adaptive encryption algorithm according to the processing efficiency requirement;

[0108] S524: Perform encryption operations on the multiple hidden calculation values ​​according to the encryption key length, encryption mode, and encryption algorithm to obtain multiple non-public scoring values.

[0109] As described in the above steps S521-S524, in the present invention, by obtaining business scenario information based on the access purpose information, it is possible to clearly understand in what business environment the hidden calculation value is used, wherein the business scenario information refers to descriptive information of various specific situations and conditions related to the access purpose. For example, if the access purpose is to conduct a cooperative credit assessment between enterprises, the business scenario may be in the business cooperation negotiation stage; if the access purpose is credit approval by a financial institution, the business scenario is in the credit application process. Different business scenarios have different requirements for data security, resource usage and processing speed. For example, business scenarios involving financial transactions usually have extremely high security level requirements, while some data analysis scenarios with low real-time requirements may pay more attention to resource constraints. Then, based on the business scenario information, encryption requirement information is obtained, which can formulate more accurate strategies for subsequent encryption operations to avoid over-encryption or under-encryption. Among them, encryption requirement information refers to a series of requirements of small and micro enterprises in data encryption, including security level requirements, that is, the degree of data confidentiality that the enterprise expects to achieve. A high security level means stricter requirements for data confidentiality; resource constraint requirements, that is, the requirements of the enterprise in computing resources and storage resources. Restrictions in other aspects will affect the selection of encryption schemes; processing efficiency requirements, which reflect the enterprise's expectations for the completion time of encryption operations. Based on the security level requirements, the encryption key length should be reasonably determined. The encryption key length refers to the number of binary bits contained in the encryption key. Generally speaking, the higher the security level, the longer the key length is required to enhance encryption security. Based on the resource constraints of small and micro enterprises, appropriate encryption modes should be selected. The encryption mode refers to the method and rules for encrypting data. For example, for small and micro enterprises with limited resources, lightweight encryption modes may be selected to reduce the use of system resources. The encryption algorithm is selected based on processing efficiency requirements. The encryption algorithm refers to the specific mathematical algorithm that implements data encryption and decryption operations. The encryption algorithm can meet the enterprise's requirements for data encryption speed, prevent encryption operations from becoming a bottleneck in business processes, and ensure the smoothness of the entire credit analysis process. Then, using the determined encryption key length, encryption mode, and encryption algorithm, the calculation result information is encrypted, thereby converting the computationally valuable calculation results into non-public scoring values, ensuring the security of data during storage and transmission, and preventing data from being obtained and understood by unauthorized entities.

[0110] This application also provides a cloud computing-based small and micro enterprise credit data analysis method system, including:

[0111] A first acquisition module is used to obtain the identity information of the visitor, and obtain the visit purpose information and access permission information based on the identity information;

[0112] a second acquisition module, configured to acquire evaluation core information according to the access purpose information, wherein the evaluation core information includes evaluation model information and a plurality of core data information, and acquire an information sensitivity matching table according to the plurality of core data information;

[0113] a screening module, configured to obtain an access sensitivity threshold of the visitor based on the access permission information, and screen multiple core data information based on the access sensitivity threshold and the information sensitivity matching table to obtain corresponding multiple matching public information and multiple hidden processing information;

[0114] a first calculation module, configured to obtain a first calculation model based on the evaluation model information and a plurality of matching public information, and obtain a first credit score value based on the first calculation model and the plurality of matching public information;

[0115] A second calculation module is configured to transmit the plurality of hidden processing information to the node container for calculation according to the evaluation model information to obtain a hidden calculation value, and encrypt the hidden calculation value according to the access purpose information to obtain a non-public rating value;

[0116] a decryption module, configured to obtain a decryption key based on the identity information, and obtain a second credit score value based on the decryption key and the non-public score value;

[0117] The third calculation module is configured to obtain a credit weighting coefficient according to the access sensitivity threshold, and obtain a total credit score value according to the credit weighting coefficient, the first credit score value, and the second credit score value.

[0118] In one embodiment, the second acquisition module includes:

[0119] A first acquiring unit is configured to acquire corresponding historical collection information and information update period according to the plurality of core data information, wherein the historical collection information includes historical access frequency, visitor type frequency, and business process association frequency;

[0120] A first scoring unit is configured to obtain a collection frequency value based on the historical access frequency, visitor type frequency, and business process association frequency, and score multiple core data information based on the collection frequency value to obtain a corresponding confidentiality score value;

[0121] A second scoring unit is used to score the multiple core data information according to the information update cycle to obtain a timeliness score value;

[0122] The classification unit is used to obtain sensitivity values ​​corresponding to multiple core data information according to the confidentiality score value and the timeliness score value, and classify the multiple core data information according to the sensitivity values ​​to obtain an information sensitivity matching table.

[0123] The present invention also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned cloud computing-based small and micro enterprise credit data analysis method are implemented.

[0124] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned cloud computing-based small and micro enterprise credit data analysis method.

[0125] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0126] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0127] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A cloud computing-based small and micro enterprise credit data analysis method, characterized in that: include: Obtaining the visitor's identity information, and obtaining the visit purpose information and access permission information based on the identity information; Acquiring evaluation core information according to the access purpose information, wherein the evaluation core information includes evaluation model information and multiple core data information, and acquiring an information sensitivity matching table according to the multiple core data information; Obtaining the access sensitivity threshold of the visitor according to the access permission information, and screening multiple core data information according to the access sensitivity threshold and the information sensitivity matching table to obtain corresponding multiple matching public information and multiple hidden processing information; Obtaining a first calculation model based on the evaluation model information and a plurality of matching public information, and obtaining a first credit score value based on the first calculation model and the plurality of matching public information; Transmitting multiple hidden processing information to the node container for calculation according to the evaluation model information to obtain a hidden calculation value, and encrypting the hidden calculation value according to the access purpose information to obtain a non-public rating value; Obtaining a decryption key based on the identity information, and obtaining a second credit score value based on the decryption key and the non-public score value; A credit weighting coefficient is obtained according to the access sensitivity threshold, and a total credit score value is obtained according to the credit weighting coefficient, the first credit score value, and the second credit score value.

2. The cloud computing-based small and micro enterprise credit data analysis method according to claim 1 is characterized in that: The step of obtaining an information sensitivity matching table according to the plurality of core data information includes: Acquire corresponding historical collection information and information update cycle according to the plurality of core data information, wherein the historical collection information includes historical access frequency, visitor type frequency, and business process association frequency; Obtaining a collection frequency value based on the historical access frequency, visitor type frequency, and business process association frequency, and scoring multiple core data information based on the collection frequency value to obtain a corresponding confidentiality score value; Scoring multiple core data information according to the information update cycle to obtain a timeliness score value; The sensitivity values ​​corresponding to the multiple core data information are obtained according to the confidentiality score value and the timeliness score value, and the multiple core data information are classified according to the sensitivity values ​​to obtain an information sensitivity matching table.

3. The cloud computing-based small and micro enterprise credit data analysis method according to claim 1 is characterized in that: The step of obtaining the visitor's access sensitivity threshold according to the access permission information includes: Acquiring operation authority information according to the access authority information, wherein the operation authority information includes information reading, information editing, and information approval; Determine each operation permission according to the operation permission information, obtain operation result information, and obtain a corresponding sensitivity threshold interval according to the operation result information; Acquiring operation detail information according to the operation authority information, wherein the operation detail information includes operation duration and operation location, and acquiring a first weight coefficient according to the operation duration; Obtaining a location value according to the operation location, and obtaining a second weight coefficient according to the location value; An access sensitivity threshold is obtained according to the sensitivity threshold interval, the operation duration, the first weight coefficient, the location assignment, and the second weight coefficient.

4. The cloud computing-based small and micro enterprise credit data analysis method according to claim 1 is characterized in that: The step of obtaining a first calculation model based on the evaluation model information and a plurality of matching public information, and obtaining a first credit score value based on the first calculation model and the plurality of matching public information, includes: Acquire evaluation dimension information and evaluation logic information according to the evaluation model information, and acquire multiple key factor information according to the evaluation dimension information and multiple matching public information; Obtaining key sensitivity values ​​corresponding to the plurality of key factor information, and prioritizing the plurality of key factor information according to the key sensitivity values ​​to obtain key factor ranking information, and obtaining indicator weight parameters corresponding to the key factor information according to the evaluation logic information and the key factor ranking information; Constructing a first calculation model according to the evaluation logic information and indicator weight parameters; Corresponding key factor values ​​are obtained according to the plurality of key factor information, and a first credit score value is obtained according to the plurality of key factor values ​​and a first calculation model.

5. The cloud computing-based small and micro enterprise credit data analysis method according to claim 1 is characterized in that: The step of transmitting the plurality of hidden processing information to the node container for calculation according to the evaluation model information to obtain the hidden calculation value includes: Acquire a second calculation model according to the evaluation model information, and acquire the number of scoring factor input layers and the number of scoring associated processing layers according to the second calculation model; Acquire a first bias value and a second bias value according to the second calculation model; Acquire multiple scoring factor values ​​according to the anonymized information, and acquire corresponding first connection weights according to the scoring factor values; Obtaining a scoring factor summary value according to the scoring factor value, the first connection weight, and the first bias value; Performing a linear transformation on the summary value of the scoring factors to obtain an input value of the scoring association processing layer; A corresponding second connection weight is obtained according to the score-associated processing layer input value, and a hidden calculation value is obtained according to the score-associated processing layer input value, the second connection weight and the second bias value.

6. The cloud computing-based small and micro enterprise credit data analysis method according to claim 1 is characterized in that: The step of encrypting the hidden calculated value according to the access purpose information to obtain a non-public rating value includes: Acquiring business scenario information according to the access purpose information, and acquiring encryption requirement information according to the business scenario information, wherein the encryption requirement information includes security level requirements, resource limitation requirements, and processing efficiency requirements; Acquire an encryption key length that matches the security level requirement, and acquire an adaptive encryption mode based on the resource limitation requirement; Obtaining an adapted encryption algorithm according to the processing efficiency requirement; The multiple hidden calculation values ​​are encrypted according to the encryption key length, encryption mode and encryption algorithm to obtain multiple non-public scoring values.

7. A cloud computing-based small and micro enterprise credit data analysis system, characterized by: include: A first acquisition module is used to obtain the identity information of the visitor, and obtain the visit purpose information and access permission information based on the identity information; a second acquisition module, configured to acquire evaluation core information according to the access purpose information, wherein the evaluation core information includes evaluation model information and a plurality of core data information, and acquire an information sensitivity matching table according to the plurality of core data information; a screening module, configured to obtain an access sensitivity threshold of the visitor based on the access permission information, and screen multiple core data information based on the access sensitivity threshold and the information sensitivity matching table to obtain corresponding multiple matching public information and multiple hidden processing information; a first calculation module, configured to obtain a first calculation model based on the evaluation model information and a plurality of matching public information, and obtain a first credit score value based on the first calculation model and the plurality of matching public information; A second calculation module is configured to transmit the plurality of hidden processing information to the node container for calculation according to the evaluation model information to obtain a hidden calculation value, and encrypt the hidden calculation value according to the access purpose information to obtain a non-public rating value; a decryption module, configured to obtain a decryption key based on the identity information, and obtain a second credit score value based on the decryption key and the non-public score value; The third calculation module is configured to obtain a credit weighting coefficient according to the access sensitivity threshold, and obtain a total credit score value according to the credit weighting coefficient, the first credit score value, and the second credit score value.

8. The cloud computing-based small and micro enterprise credit data analysis system according to claim 7 is characterized in that: The second acquisition module includes: A first acquiring unit is configured to acquire corresponding historical collection information and information update period according to the plurality of core data information, wherein the historical collection information includes historical access frequency, visitor type frequency, and business process association frequency; A first scoring unit is configured to obtain a collection frequency value based on the historical access frequency, visitor type frequency, and business process association frequency, and score multiple core data information based on the collection frequency value to obtain a corresponding confidentiality score value; A second scoring unit is used to score the multiple core data information according to the information update cycle to obtain a timeliness score value; The classification unit is used to obtain sensitivity values ​​corresponding to multiple core data information according to the confidentiality score value and the timeliness score value, and classify the multiple core data information according to the sensitivity values ​​to obtain an information sensitivity matching table.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images