Method and system for automatic compliance verification of title declaration materials

Abstract

This application belongs to the field of material compliance verification and discloses an automated compliance verification method and system for professional title application materials. First, it obtains the citation relationships between multiple supporting materials for professional title applications, constructing a supporting material association network with each supporting material as a node and the citation relationships as connecting edges. Next, it obtains the material risk coefficient for each supporting material, and determines the adjustment risk coefficient for each material by combining the connecting edges with the risk coefficient of each individual material. Based on this, it summarizes the application risk coefficient for the entire batch of supporting materials. Finally, it compares the application risk coefficient with a preset threshold. If the coefficient is greater than or equal to the threshold, the batch of materials is marked as unqualified; otherwise, it is marked as qualified, effectively improving the accuracy of risk assessment for professional title application materials.

Description

Technical Field

[0001] This application relates to the field of material compliance verification technology, and more specifically, to an automated compliance verification method and system for professional title application materials. Background Technology

[0002] The review of application materials for professional titles is a core step in the professional title evaluation process. Currently, manual review is commonly used to verify these materials. Reviewers typically need to check each item submitted by the applicant against the corresponding professional title evaluation standards, including personal information, work experience, achievements, award certificates, academic achievements, and other multi-dimensional materials, to determine the authenticity and completeness of the materials. In some scenarios, the validity of the materials may also need to be manually assessed based on the specific requirements of the applicant's industry and the applied position. This review model is effective in small-scale evaluations with a limited number of applicants, relying on the reviewers' experience to complete the basic review work. It is currently a widely used and conventional technique in the field of professional title application review.

[0003] When manual review is used to examine application materials for professional titles, it is difficult to comprehensively verify cross-departmental and cross-field citations and related information. Some more subtle risks, such as falsified materials and inflated performance records, cannot be identified in a timely manner, directly leading to inaccurate risk assessment results. The work attitude and experience level of the reviewers also affect the completeness of risk identification, further reducing the accuracy of the risk assessment results and failing to provide a reliable reference for subsequent professional title evaluation. Summary of the Invention

[0004] The purpose of this application is to provide an automated compliance verification method and system for professional title application materials, which solves the technical problem of inaccurate risk assessment when manually reviewing professional title application materials, and achieves the technical effect of improving the accuracy of risk assessment when reviewing professional title application materials.

[0005] In a first aspect, embodiments of this application provide an automated compliance verification method for professional title application materials. The method includes: obtaining the citation relationships between multiple supporting materials used for professional title application; constructing a supporting material association network by treating each supporting material as a node and the citation relationships between multiple supporting materials as association edges; obtaining the material risk coefficient of each supporting material; determining the adjustment risk coefficient of each supporting material based on the material risk coefficient of each supporting material through the association edges between multiple supporting materials; determining the application risk coefficient corresponding to the multiple supporting materials based on the adjustment risk coefficients of the multiple supporting materials; marking the multiple supporting materials as unqualified when the application risk coefficient is greater than or equal to a preset application risk coefficient; and marking the multiple supporting materials as qualified when the application risk coefficient is less than the preset application risk coefficient.

[0006] In one possible implementation, the adjustment risk coefficient of each supporting material is determined based on its material risk coefficient through the association edges between multiple supporting materials. This includes: identifying the first-level associated supporting materials (associated through adjacent nodes), the second-level associated supporting materials (associated through one intermediate node), and the third-level associated supporting materials (associated through two nodes) for each supporting material through the association edges between multiple supporting materials; obtaining the first-level material risk coefficient, the second-level material risk coefficient, and the third-level material risk coefficient for each supporting material's first-level associated supporting materials; obtaining the first-level association weight, the second-level association weight, and the third-level association weight for each supporting material's second-level associated supporting materials; and determining the sum of the material risk coefficient, the product of the first-level association weight and the first-level material risk coefficient, the product of the second-level association weight and the second-level material risk coefficient, and the product of the third-level association weight and the third-level material risk coefficient for each supporting material as the adjustment risk coefficient for each supporting material.

[0007] In another possible implementation, the declaration risk coefficient corresponding to multiple supporting materials is determined based on the adjustment risk coefficient of multiple supporting materials. This includes: obtaining the supporting material weight corresponding to each supporting material; determining the sum of the products of the adjustment risk coefficient and the supporting material weight of multiple supporting materials as the aggregate declaration risk coefficient; and determining the quotient of the aggregate declaration risk coefficient divided by the number of supporting materials of the multiple supporting materials as the declaration risk coefficient corresponding to the multiple supporting materials.

[0008] In another possible implementation, the material risk coefficient for each supporting document is obtained, including: obtaining the information deviation degree, the reverse value of evidence credibility, the historical duplication rate, and the material anomaly degree for each supporting document; obtaining the weights of the information deviation degree, the reverse value of evidence credibility, the historical duplication rate, and the material anomaly degree; wherein, the information deviation degree characterizes the degree of information deviation of the supporting document, the reverse value of evidence credibility is the reciprocal of the credibility of the supporting document as supporting evidence, the historical duplication rate characterizes the duplication rate of the supporting document as a professional title application material within a preset time period, and the material anomaly degree characterizes the degree of content anomaly of the supporting document; the sum of the product of the information deviation degree and the information deviation degree weight, the product of the reverse value of evidence credibility and the reverse value of evidence credibility, the product of the historical duplication rate and the historical duplication rate weight, and the product of the material anomaly degree and the material anomaly degree weight for each supporting document is determined as the material risk coefficient for each supporting document.

[0009] In another possible implementation, the adjustment risk coefficient of each supporting material is determined based on the material risk coefficient of each material through the association edges between multiple supporting materials. This also includes: determining the number of times each supporting material is cited; determining the citation impact coefficient of each supporting material based on its citation count; obtaining the first-level citation impact coefficient, the second-level citation impact coefficient, and the third-level citation impact coefficient corresponding to the first-level related supporting materials, the second-level citation impact coefficient, and the third-level citation impact coefficient for each supporting material; determining the product of the first-level association weight, the first-level material risk coefficient, and the first-level citation impact coefficient for each supporting material as the first risk coefficient; determining the product of the second-level association weight, the second-level material risk coefficient, and the second-level citation impact coefficient for each supporting material as the second risk coefficient; determining the product of the third-level association weight, the third-level material risk coefficient, and the third-level citation impact coefficient for each supporting material as the third risk coefficient; and determining the sum of the product of the material risk coefficient and the citation impact coefficient, the first risk coefficient, the second risk coefficient, and the third risk coefficient for each supporting material as the adjustment risk coefficient for each supporting material.

[0010] In another possible implementation, the citation impact coefficient of each supporting material is determined based on the number of times each supporting material is cited. This includes: obtaining the supporting material type of each supporting material; obtaining the citation material weight corresponding to the supporting material type of each supporting material; determining the product of the number of citations of each supporting material and the citation material weight corresponding to the supporting material type as the basic citation impact coefficient; and determining the quotient of the basic citation impact coefficient of each supporting material divided by the number of supporting materials in multiple supporting materials as the citation impact coefficient of each supporting material.

[0011] In another possible implementation, the information deviation degree of each supporting material is obtained, including: extracting the material key fields and material non-key fields of each supporting material; obtaining the official key fields and official non-key fields corresponding to the material key fields of each supporting material queried from the official database; wherein, the material key fields include name, date, level, position, and number; when the material key fields and official key fields are completely consistent, and the material non-key fields and official non-key fields are completely consistent, the information deviation degree is determined to be 0; when the material key fields and official key fields are completely consistent, and the material non-key fields and official non-key fields are not completely consistent, the information deviation degree is determined to be 0.2; when the material key fields and official key fields are not completely consistent, the information deviation degree is determined to be 0.6; when no official key field corresponding to the material key fields of the supporting material is found in the official database, the information deviation degree is determined to be 0.8; when the number of inconsistencies between the material key fields and official key fields is greater than a preset number of inconsistencies, the information deviation degree is determined to be 1.

[0012] In another possible implementation, the reverse credibility value of each supporting document is obtained, including: when the supporting document is verified in an official database, the reverse credibility value is determined to be 0.1; when the supporting document is a scanned copy of the original certificate issued by an official organization, the reverse credibility value is determined to be 0.3; when the supporting document is a certificate issued by an organization and stamped with an official seal, the reverse credibility value is determined to be 0.5; when the supporting document is a document filled out by the applicant without supporting documents, the reverse credibility value is determined to be 0.8.

[0013] In another possible implementation, the historical duplication rate and material anomaly degree of each supporting document are obtained, including: retrieving the usage records of supporting documents in the historical application database; determining the historical duplication rate to be 0 when the supporting document is used for the first time; determining the historical duplication rate to be 0.2 when the same applicant has used supporting documents in different levels of historical application records and it has been marked as repeated use; determining the historical duplication rate to be 0.7 when the supporting document has been used by other applicants; determining the historical duplication rate to be 0.9 when the supporting document has been used multiple times in this application and scored repeatedly; identifying timeline crossing anomalies, numerical anomalies, and name confusion anomalies in the supporting documents; and determining the duration of the supporting document's existence. When there are timeline crossing anomalies, numerical anomalies, and homonymy anomalies, the material anomaly degree is determined to be 1; when the supporting materials only have timeline crossing anomalies and numerical anomalies, the material anomaly degree is determined to be 0.8; when the supporting materials only have timeline crossing anomalies, the material anomaly degree is determined to be 0.3; when the supporting materials have no anomalies, the material anomaly degree is determined to be 0. Among them, timeline crossing anomalies include the achievement date being earlier than the start date; numerical anomalies include discrepancies between the actual journal quartile and the journal's marked quartile in the supporting materials, and the project funding in the supporting materials being greater than or equal to the regular project funding in similar supporting materials; homonymy anomalies include supporting materials with the same author name but different author affiliations.

[0014] Secondly, embodiments of this application provide an automated compliance verification system for professional title application materials, including units for implementing the above-described method.

[0015] The beneficial effects of the embodiments in this application compared with the prior art are: This application provides an automated compliance verification method for professional title application materials. The method includes: obtaining the citation relationships between multiple supporting materials used for professional title application; constructing a supporting material association network by treating each supporting material as a node and the citation relationships between multiple supporting materials as association edges; obtaining the material risk coefficient of each supporting material; determining the adjustment risk coefficient of each supporting material based on the material risk coefficient of each supporting material through the association edges between multiple supporting materials; determining the application risk coefficient corresponding to multiple supporting materials based on the adjustment risk coefficients of multiple supporting materials; marking multiple supporting materials as unqualified when the application risk coefficient is greater than or equal to a preset application risk coefficient; and marking multiple supporting materials as qualified when the application risk coefficient is less than the preset application risk coefficient. In this embodiment, the citation relationships between multiple supporting materials used for professional title applications are obtained. Then, a supporting material association network is constructed with each supporting material as a node and the citation relationship as an associated edge. The material risk coefficient of each supporting material is obtained. Then, by combining the associated edges of the supporting material association network with the material risk coefficient of each supporting material, the adjustment risk coefficient of each supporting material is calculated. This can incorporate the risk transmission effect of associated materials into the calculation, making the risk assessment results more consistent with the actual associated risk situation of the materials and improving the accuracy of risk assessment. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating the first automated compliance verification method for professional title application materials provided in this application embodiment; Figure 2 A schematic diagram illustrating the workflow of the first automated compliance verification method for professional title application materials provided in this application embodiment; Figure 3 A flowchart illustrating the second automated compliance verification method for professional title application materials provided in this application embodiment; Figure 4 A schematic diagram illustrating the workflow of the second automated compliance verification method for professional title application materials provided in this application embodiment; Figure 5 A flowchart illustrating the third automated compliance verification method for professional title application materials provided in this application embodiment; Figure 6A schematic diagram illustrating the workflow of the third automated compliance verification method for professional title application materials provided in this application embodiment; Figure 7 A flowchart illustrating the fourth automated compliance verification method for professional title application materials provided in this application embodiment; Figure 8 A schematic diagram of the workflow for the fourth automated compliance verification method for professional title application materials provided in this application embodiment; Figure 9 A flowchart illustrating the fifth automated compliance verification method for professional title application materials provided in this application embodiment; Figure 10 A schematic diagram of the workflow for the fifth automated compliance verification method for professional title application materials provided in this application embodiment; Figure 11 A flowchart illustrating the sixth automated compliance verification method for professional title application materials provided in this application embodiment; Figure 12 A schematic diagram illustrating the workflow of the sixth automated compliance verification method for professional title application materials provided in this application embodiment; Figure 13 This is a schematic diagram of an automated compliance verification system for professional title application materials provided in an embodiment of this application. Detailed Implementation

[0018] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0019] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0020] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0021] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0022] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0023] The risk assessment of related supporting materials is inaccurate when manually reviewing application materials for professional titles.

[0024] Based on the above reasons, this application provides an automated compliance verification method for professional title application materials. The method includes: obtaining the citation relationships between multiple supporting materials used for professional title application; constructing a supporting material association network by treating each supporting material as a node and the citation relationships between multiple supporting materials as association edges; obtaining the material risk coefficient of each supporting material; determining the adjustment risk coefficient of each supporting material based on the material risk coefficient of each supporting material through the association edges between multiple supporting materials; determining the application risk coefficient corresponding to multiple supporting materials based on the adjustment risk coefficients of multiple supporting materials; marking multiple supporting materials as unqualified when the application risk coefficient is greater than or equal to a preset application risk coefficient; and marking multiple supporting materials as qualified when the application risk coefficient is less than the preset application risk coefficient. In this embodiment, the citation relationships between multiple supporting materials used for professional title applications are obtained. Then, a supporting material association network is constructed with each supporting material as a node and the citation relationship as an associated edge. The material risk coefficient of each supporting material is obtained. Then, by combining the associated edges of the supporting material association network with the material risk coefficient of each supporting material, the adjustment risk coefficient of each supporting material is calculated. This can incorporate the risk transmission effect of associated materials into the calculation, making the risk assessment results more consistent with the actual associated risk situation of the materials and improving the accuracy of risk assessment.

[0025] In some scenarios, the automated compliance verification method for professional title application materials according to the embodiments of this application can be applied to the centralized application scenario of professional titles in public institutions, and batch verify the association compliance of papers, projects, award certificates and other materials of multiple batches of applicants, thereby reducing the false negative rate of manual review.

[0026] In other scenarios, the automated compliance verification method for professional title application materials according to the embodiments of this application can also be applied to the scenario of applicant self-review. After uploading materials, applicants can receive risk warnings in real time, correct material problems in advance, and improve the first-time pass rate of the application.

[0027] The following section provides a detailed explanation of an automated compliance verification method for professional title application materials provided in this application, using specific examples.

[0028] Figure 1 A flowchart illustrating the first automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 1 As shown in the embodiment of this application, an automated compliance verification method for professional title application materials is provided. The method includes S110 to S120, and S110 to S120 are described in detail below.

[0029] S110. Obtain the citation relationships between multiple supporting documents used for professional title application. Treat each supporting document as a node and the citation relationships between multiple supporting documents as association edges to construct a supporting document association network.

[0030] Figure 2 A schematic diagram of the workflow of the first automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 2 As shown in this implementation method, all supporting materials to be submitted during the professional title application process can be sorted out first, and each independent supporting material can be extracted separately as a basic component of the subsequent network, providing basic node support for the subsequent construction of the network.

[0031] For example, applicants can organize all the materials they need to submit for professional title evaluation, classify and store each material separately, and label the name and category of each material to complete the basic sorting of all nodes.

[0032] In this implementation, the logical relationships of citation between different supporting materials can be sorted out, and an association edge can be added between each pair of supporting materials that have a citation relationship. The citation support logic between the materials can be reflected through the association edge.

[0033] For example, applicants can compare the relationship between each pair of materials one by one, identify material pairs that are supported by citations, mark the citation direction and relationship between the corresponding materials, and complete the sorting out of all related edges.

[0034] In this implementation method, the nodes and reference relationships of the supporting materials that have been sorted out can be integrated, and a supporting material relationship network can be built according to the correspondence between nodes and edges. This intuitively presents the relationship logic of all materials, which not only makes it convenient for applicants to check the completeness of their materials, but also makes it convenient for reviewers to quickly sort out the review logic.

[0035] For example, applicants can use knowledge graph tools to import all the labeled supporting material nodes into the tool, and then import the organized association edge information, so that the tool can automatically generate a visual network of supporting material associations.

[0036] For example, common types of supporting materials for professional title applications include basic information materials, performance achievement materials, academic achievement materials, continuing education materials, and honors and awards materials. Basic information materials prove the applicant's basic identity and qualifications; performance achievement materials prove the applicant's work performance; academic achievement materials prove the applicant's academic research output; continuing education materials prove the applicant has completed the required continuing education hours for the professional title; and honors and awards materials prove the applicant has received officially recognized awards.

[0037] For example, common types of citation relationships among multiple supporting materials for professional title applications include supporting correspondence, source citation, and supplementary explanation. Supporting correspondence means that one material is used to prove the authenticity of the application information in another material; source citation means that the content of one achievement material cites the content of another material; and supplementary explanation means that one material supplements and explains the incomplete content in another material.

[0038] For example, the process of constructing a supporting materials association network can be completed according to the following steps: Step 1: Organize all supporting materials and extract nodes: Applicants for senior professional titles should organize 24 materials, including their personal ID card, professional title qualification certificate, three published papers, five project acceptance reports, ten continuing education completion certificates, and three personal honorary certificates. Each material serves as an independent node. Step 2: Organize all citation relationships and extract related edges: The applicant's information as the project leader in the project acceptance report is supported by the qualification certificate, generating a related edge from the qualification certificate to the corresponding project acceptance report. If the paper content cites experimental data from the project research process, and the corresponding data is in the project acceptance report, a related edge from the project acceptance report to the corresponding paper is generated. If a project acceptance report does not specify the source of project funding, the applicant should upload the project approval document, which provides supplementary explanation to the project acceptance report, generating a related edge from the project approval document to the corresponding project acceptance report. Step 3: Integrate nodes and related edges to generate the network: Import all nodes and related edges into a visualization tool to automatically generate a complete supporting materials association network.

[0039] S120. Obtain the material risk coefficient for each supporting document. Based on the material risk coefficient of each supporting document and the association edges between multiple supporting documents, determine the adjustment risk coefficient for each supporting document. Based on the adjustment risk coefficients of multiple supporting documents, determine the corresponding declaration risk coefficient for each supporting document. When the declaration risk coefficient is greater than or equal to the preset declaration risk coefficient, mark multiple supporting documents as unqualified. When the declaration risk coefficient is less than the preset declaration risk coefficient, mark multiple supporting documents as qualified.

[0040] In this implementation method, for each piece of supporting material that has been prepared, the material risk coefficient of the corresponding single material can be calculated in combination with the preset verification rules, so as to provide basic data for subsequent calculation of adjustment risk coefficient.

[0041] For example, for each supporting document uploaded to the professional title application system, the system can perform authenticity verification, format verification, and timeliness verification, and then generate a material risk coefficient for each document and store it in the corresponding database.

[0042] In this implementation, the material risk coefficients of adjacent nodes can be transmitted and integrated according to certain rules based on the established supporting material association network and the reference relationship corresponding to the associated edges. The original material risk coefficient of each node can be adjusted to obtain the adjusted risk coefficient of each supporting material.

[0043] For example, each association edge in the supporting material association network can be traversed, and the material risk coefficient of the cited material pointed to by the reference can be superimposed on the original material risk coefficient of the cited material according to the weight ratio of the reference relationship, thereby completing the adjustment of the original material risk coefficient and obtaining the corresponding adjusted risk coefficient.

[0044] In this implementation method, the adjustment risk coefficients of all supporting materials can be integrated and calculated, and the application risk coefficient corresponding to the entire application materials can be obtained by combining the risk weights of all materials, thereby realizing a quantitative assessment of the overall risk of the application materials.

[0045] For example, the adjustment risk coefficients of different categories of supporting materials can be multiplied by the corresponding category weights, and then all weighted adjustment risk coefficients can be summed to obtain the final declaration risk coefficient.

[0046] In this implementation method, the calculated declaration risk coefficient can be compared with the preset declaration risk coefficient. When the value of the declaration risk coefficient is greater than or equal to the preset declaration risk coefficient, the unqualified marking operation is performed to clearly indicate that the supporting materials for this batch of declarations are unqualified as a whole.

[0047] For example, if the preset risk coefficient for declaration is 0.6 and the calculated risk coefficient is 0.7, which meets the conditions for triggering the marking of non-compliance, the system can mark all of the applicant's multiple supporting materials as non-compliance.

[0048] For example, when supporting materials are marked as unqualified, the system can first generate an unqualified prompt on the applicant's personal application portal, mark the core related material group with the risk coefficient exceeding the standard, and then simultaneously send the marking result to the review system to save the unqualified marking record.

[0049] For example, if an applicant's application risk coefficient is 0.65, which is greater than the preset application risk coefficient of 0.6, the system will automatically generate a prompt in the applicant's background that "the overall risk of the application materials does not meet the standard and the materials are marked as unqualified". At the same time, it will mark the papers with high risk and the corresponding citations, thus completing the unqualified marking process.

[0050] In this implementation method, the calculated declaration risk coefficient can be compared with the preset declaration risk coefficient. When the value of the declaration risk coefficient is less than the preset declaration risk coefficient, the qualified marking operation is performed to clearly indicate that the supporting materials for this batch of declarations are qualified as a whole.

[0051] For example, if the preset risk coefficient for declaration is 0.6 and the calculated risk coefficient is 0.42, which meets the conditions for triggering the qualified marking, the system can mark all of the applicant's multiple supporting materials as qualified.

[0052] For example, when the supporting materials are marked as qualified, the system can first generate a qualified prompt on the applicant's personal application portal to inform the applicant that the material risk check has passed, and then simultaneously synchronize the qualified marking result to the review system and store it in the corresponding review queue.

[0053] For example, if an applicant's risk coefficient is 0.38, which is less than the preset risk coefficient of 0.6, the system will automatically generate a message in the applicant's background that "the risk check of the application materials has passed and the materials are marked as qualified", and move the application materials into the review queue to complete the qualified marking process.

[0054] This implementation method obtains the citation relationships between multiple supporting materials used for professional title applications. By constructing a supporting material association network with each supporting material as a node and the citation relationship as an associated edge, it can completely restore the internal association logic of all supporting materials, avoid missing association contradictions when verifying a single material in isolation, and improve the comprehensiveness of verification.

[0055] This implementation method obtains the material risk coefficient of each supporting material, and then calculates the adjusted risk coefficient of each supporting material by combining the material risk coefficient of each supporting material with the associated edges of the supporting material association network. This can incorporate the risk transmission effect of associated materials into the calculation, making the risk assessment results more consistent with the actual associated risk situation of the materials and improving the accuracy of risk assessment.

[0056] This method calculates the corresponding application risk coefficient based on the adjustment risk coefficient of multiple supporting materials, and then compares the application risk coefficient with the preset application risk coefficient to mark the supporting materials as qualified or unqualified in batches. This eliminates the need for manual verification of multiple related materials one by one, greatly reducing the labor cost of verifying professional title application materials and improving the overall efficiency of verification.

[0057] Figure 3 A flowchart illustrating the second automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 3 As shown, in some implementations, in S120 above, the adjustment risk coefficient of each supporting material is determined based on the material risk coefficient of each supporting material through the association edge between multiple supporting materials, including S121 to S122. S121 to S122 will be explained in detail below.

[0058] S121. Through the association edges between multiple supporting materials, determine the first-level associated supporting materials for each supporting material through adjacent nodes, the second-level associated supporting materials through one intermediate node, and the third-level associated supporting materials through two nodes. Obtain the first-level material risk coefficient, the second-level material risk coefficient, and the third-level material risk coefficient corresponding to the first-level associated supporting materials for each supporting material.

[0059] Figure 4 A schematic diagram of the workflow for the second automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 4 As shown, in this implementation, the existing supporting material association network can be relied upon to traverse all the association edges corresponding to each target supporting material. The association materials can be hierarchically divided according to the number of nodes passed through during the association edge transmission process, distinguishing between first-level, second-level, and third-level supporting materials.

[0060] For example, starting with the target supporting material, adjacent nodes are selected from the nodes directly connected by the associated edges to obtain first-level associated supporting material. The target node itself is excluded from the adjacent nodes of the first-level associated nodes to obtain second-level associated supporting material connected by one intermediate node. Then, the previously selected first-level and target nodes are excluded from the adjacent nodes of the second-level associated nodes to obtain third-level associated supporting material connected by two intermediate nodes.

[0061] For example, taking the target paper node as an example, the target paper directly quotes the content of the project acceptance report, and the two are directly connected by an edge. The project acceptance report is the primary supporting evidence for the target paper. The project acceptance report was undertaken by the applicant during their tenure, and a qualification certificate is required to prove their position. The qualification certificate is connected to the target paper through the project acceptance report, passing through one node (the project acceptance report). The qualification certificate is the secondary supporting evidence for the target paper. The qualification certificate is issued by the local human resources and social security department, and an official document from the department is required to prove its issuing qualification. The official document is connected to the target paper through the qualification certificate and the project acceptance report, passing through two nodes. The official document is the tertiary supporting evidence for the target paper.

[0062] In this implementation, the risk coefficients of all related supporting materials under the corresponding level can be extracted based on the hierarchical division results, and the risk coefficient set of the corresponding level can be obtained, providing hierarchical data support for subsequent risk transmission and superposition calculations.

[0063] For example, after completing the hierarchical division of the target supporting materials, the original risk coefficients of all first-level related supporting materials are extracted and organized into a set of first-level material risk coefficients. The original risk coefficients of all second-level related supporting materials are extracted and organized into a set of second-level material risk coefficients. The original risk coefficients of all third-level related supporting materials are extracted and organized into a set of third-level material risk coefficients.

[0064] For example, based on the above-mentioned division of the target paper nodes, the original risk coefficient of the project acceptance report is 0.3, corresponding to a Level 1 material risk coefficient of 0.3. The original risk coefficient of the professional qualification certificate is 0.1, corresponding to a Level 2 material risk coefficient of 0.1. The original risk coefficient of the official document issued by the human resources and social security department is 0.05, corresponding to a Level 3 material risk coefficient of 0.05. If the target paper has multiple supporting materials at the same level, the risk coefficients corresponding to each supporting material can be listed sequentially.

[0065] S122. Obtain the first-level association weights, second-level association weights, and third-level association weights corresponding to the first-level association supporting materials. Determine the sum of the material risk coefficient, the product of the first-level association weight and the first-level material risk coefficient, the product of the second-level association weight and the second-level material risk coefficient, and the product of the third-level association weight and the third-level material risk coefficient for each supporting material, as the adjusted risk coefficient for each supporting material.

[0066] In this implementation, the pre-set weight values ​​of the corresponding levels can be matched with the first-level, second-level, and third-level supporting materials that have been divided to obtain the association weights corresponding to each level, providing weight parameters for subsequent weighted summation calculations.

[0067] For example, based on the hierarchical division of the network of supporting materials for professional title applications, the system can directly retrieve the preset first-level association weights and assign them to all first-level related supporting materials, assign second-level association weights to all second-level related supporting materials, and assign third-level association weights to all third-level related supporting materials.

[0068] For example, the weights for first-level, second-level, and third-level associations are set based on factors such as the distance between the associations, the strength of the reference relationship, and the degree of risk impact. The closer the associations are, the stronger the reference relationship is, and the greater the degree of risk impact, the higher the corresponding weight values ​​are set.

[0069] For example, first-level supporting materials are directly related to the target material, exhibiting the strongest citation relationship, and are assigned a weight of 0.8. Second-level supporting materials are related through an intermediate node, exhibiting the next strongest citation relationship, and are assigned a weight of 0.4. Third-level supporting materials are related through two intermediate nodes, exhibiting the weakest citation relationship, and are assigned a weight of 0.1.

[0070] In this implementation, the material risk coefficient of the target supporting material can be calculated by adding the product of the first-level correlation weight and the corresponding first-level material risk coefficient, the product of the second-level correlation weight and the corresponding second-level material risk coefficient, and the product of the third-level correlation weight and the corresponding third-level material risk coefficient, according to the weighted summation calculation rules. The final sum is used as the adjustment risk coefficient of the target supporting material.

[0071] For example, for supporting materials for a certain target, after compiling the risk coefficient of the materials themselves, the corresponding hierarchical correlation weight, and the risk coefficient of the materials at the corresponding level, the final adjusted risk coefficient is obtained by calculating according to the weighted summation formula.

[0072] For example, the risk coefficient of a target paper's own materials is 0.2. The first-level correlation weight of the project acceptance report, which is the primary supporting material, is 0.8, and the risk coefficient of the primary materials is 0.3. The second-level correlation weight of the qualification certificate, which is the secondary supporting material, is 0.4, and the risk coefficient of the secondary materials is 0.1. The third-level correlation weight of the official document, which is the tertiary supporting material, is 0.1, and the risk coefficient of the tertiary materials is 0.05. The calculation process is 0.2 + (0.8 × 0.3) + (0.4 × 0.1) + (0.1 × 0.05) = 0.2 + 0.24 + 0.04 + 0.005 = 0.485. Finally, the adjustment risk coefficient of the target paper is 0.485.

[0073] This implementation method divides each piece of supporting evidence into first-level, second-level, and third-level related supporting evidence based on the association edges of the supporting evidence association network. This can accurately distinguish the degree of association between different supporting evidence and the target supporting evidence, avoid interference from materials with different association distances on risk calculation, and ensure the rationality of risk stratification.

[0074] This implementation method matches primary, secondary, and tertiary association weights to supporting materials at different levels, assigning differentiated impact weights according to the degree of association. This aligns with the real-world pattern that supporting materials with closer associations have stronger risk transmission, thus improving the accuracy of risk coefficient calculation.

[0075] This implementation method sums the product of the material risk coefficient of the supporting materials themselves and the risk coefficients and corresponding weights of the supporting materials at each level, and obtains the adjusted risk coefficient. This can cover the risk of the materials themselves and the risk of transmission through multiple levels, avoiding the bias of only considering the risk of a single source, and making the risk assessment results more in line with the actual situation.

[0076] Figure 5 A flowchart illustrating the third automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 5 As shown, in some implementation methods, in the above-mentioned S120, the declaration risk coefficient corresponding to multiple supporting materials is determined based on the adjustment risk coefficient of multiple supporting materials, including S123 to S124. S123 to S124 will be explained in detail below.

[0077] S123. Obtain the weight of each supporting material.

[0078] Figure 6 A schematic diagram of the workflow for the third automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 6As shown, in this implementation method, the weight of each supporting document can be obtained by matching a pre-set weight value with the category to which each supporting document belongs and the importance of that category in the professional title evaluation, thus providing a weighting parameter for the subsequent calculation of the overall application risk coefficient.

[0079] For example, the system can pre-set corresponding weights according to the common categories of supporting materials for professional title applications. After the user uploads all supporting materials and completes the classification, the system automatically matches the corresponding category weight value for each supporting material, thus obtaining the weight of the supporting materials.

[0080] For example, the criteria for setting the weight of supporting materials include the weight percentage of the corresponding review indicators, the degree of influence of the materials on the applicant's eligibility, and the degree of influence of the materials on the review results. The higher the weight percentage of the corresponding review indicators, the greater the impact on the applicant's eligibility and the review results, and the higher the weight of the supporting materials should be.

[0081] For example, supporting materials for performance achievements and academic achievements correspond to the core indicators for senior professional title evaluation, with a weight of 0.3 for each. Supporting materials for basic information determine eligibility for application, with a weight of 0.15. Supporting materials for continuing education are basic requirements, with a weight of 0.1. Supporting materials for honors and awards are bonus points, with a weight of 0.15.

[0082] For example, when obtaining the weight of the corresponding supporting material, the category of each uploaded supporting material is first identified. After identifying the specific category to which the material belongs, the pre-set weight value of the category is retrieved, and the weight value is bound and stored with the supporting material to complete the acquisition process.

[0083] For example, when an applicant uploads a project acceptance report, the system identifies the material as supporting material for performance achievements, directly retrieves the pre-set weight of 0.3 for performance achievements, and binds the weight of 0.3 to the project acceptance report, thus completing the process of obtaining the weight of the corresponding supporting material.

[0084] S124. The sum of the products of the adjustment risk coefficients and the weights of the supporting documents for multiple supporting materials shall be used as the aggregated declaration risk coefficient. The quotient of the aggregated declaration risk coefficient divided by the number of supporting documents for multiple supporting materials shall be used as the declaration risk coefficient for each of the multiple supporting documents.

[0085] In existing technologies, when calculating the overall risk of professional title application, the weights are not allocated based on the importance of supporting materials, nor is the impact of the difference in the number of supporting materials on the overall risk calculation results eliminated. This can easily lead to interference in the final risk assessment due to the quantity of supporting materials uploaded, making it impossible to obtain an accurate application risk coefficient.

[0086] In this implementation, all supporting materials that have obtained the adjusted risk coefficient and corresponding supporting material weight can be traversed. The product of the adjusted risk coefficient and the supporting material weight of each supporting material can be calculated in turn. Then, all the product results are added together, and the sum is used as the aggregated declaration risk coefficient, which provides the basic data for the subsequent calculation of the final declaration risk coefficient.

[0087] For example, the system can traverse all supporting materials stored in the database, extract the corresponding adjustment risk coefficient and supporting material weight one by one, and after completing the product calculation of a single material, accumulate the product result into the summary variable. After the traversal is completed, the value of the summary variable is the summary declaration risk coefficient.

[0088] For example, an applicant uploads three supporting documents. The first document has an adjustment risk coefficient of 0.485, and the weight of the supporting document is 0.3, resulting in a product of 0.485 × 0.3 = 0.1455. The second document has an adjustment risk coefficient of 0.1, and the weight of the supporting document is 0.15, resulting in a product of 0.1 × 0.15 = 0.015. The third document has an adjustment risk coefficient of 0.05, and the weight of the supporting document is 0.1, resulting in a product of 0.05 × 0.1 = 0.005. The sum of these three products is 0.1455 + 0.015 + 0.005 = 0.1655, which is the total application risk coefficient.

[0089] In this implementation, the total number of all uploaded supporting materials can be counted, and the summative declaration risk coefficient can be divided by the total number of supporting materials. The quotient is the declaration risk coefficient corresponding to all multiple supporting materials, thus eliminating the interference of the difference in the number of supporting materials on risk judgment and obtaining a standardized overall risk result.

[0090] For example, after calculating the total number of all supporting materials uploaded by the applicant, the previously calculated summary application risk coefficient is extracted, and a division is performed to obtain the quotient. This quotient is the final application risk coefficient and is stored in the corresponding application file.

[0091] For example, taking the case where the total risk coefficient of the application is calculated to be 0.1655, the applicant uploaded a total of 3 supporting documents. Dividing 0.1655 by 3, the quotient is approximately 0.055. This quotient is the application risk coefficient corresponding to the multiple supporting documents.

[0092] This approach first obtains the weight of each supporting document, which matches the importance of different supporting documents in the professional title application and assigns corresponding weights. This avoids the problem of materials of different importance having the same contribution to the overall risk calculation, thus ensuring the fairness of the risk calculation.

[0093] This implementation method calculates the sum of the products of the adjustment risk coefficients of multiple supporting materials and the weights of the supporting materials to obtain the aggregated declaration risk coefficient. It can integrate the adjustment risks of all materials based on their importance, fully cover the risk contribution of all materials, and avoid the risk impact of omitting important materials.

[0094] By dividing the total application risk coefficient by the number of supporting materials to obtain the application risk coefficient, the interference of differences in the number of materials on the overall risk assessment can be eliminated, making the risk assessment results of multiple supporting materials from different applicants comparable and improving the universality of the verification results.

[0095] Figure 7 A flowchart illustrating the fourth automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 7 As shown, in some implementations, in S120 above, the material risk coefficient of each supporting material is obtained, including S125 to S126. S125 to S126 will be explained in detail below.

[0096] S125. Obtain the information deviation degree, evidence credibility reverse value, historical duplication rate, and material anomaly degree for each supporting material. Obtain the weights of information deviation degree, evidence credibility reverse value, historical duplication rate, and material anomaly degree. Among them, information deviation degree represents the degree of information deviation of the supporting materials, evidence credibility reverse value is the reciprocal of the credibility of the supporting materials as supporting evidence, historical duplication rate represents the duplication rate of the supporting materials as professional title application materials within a preset time period, and material anomaly degree represents the degree of content anomaly of the supporting materials.

[0097] Figure 8 A schematic diagram of the workflow for the fourth automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 8 As shown, in this implementation method, risk assessment can be performed on a single piece of supporting material from four different dimensions. Risk indicators corresponding to each dimension are extracted to obtain the information deviation degree, evidence credibility reverse value, historical duplication rate and material abnormality degree of each supporting material. The four indicators correspond to the risk level of different dimensions, providing multi-dimensional basic data for subsequent calculation of material risk coefficient.

[0098] For example, for a single uploaded supporting document, the system analyzes it from four dimensions: information matching degree, evidence credibility, historical declaration status, and content compliance, extracts the corresponding four risk indicators, and stores them in the information archive of the corresponding document.

[0099] In this implementation, corresponding dimension weights can be pre-set for four different risk dimensions, and the weight values ​​for each dimension can be obtained to provide weight parameters for subsequent weighted calculation of material risk coefficients.

[0100] For example, the four weights are set based on factors such as the degree of influence of the dimension on material risk, the warning priority of the risk in that dimension, and the degree of importance attached to the risk in that dimension by the professional title evaluation. The greater the influence of a certain dimension on material risk, the higher the warning priority, and the higher the corresponding weight value.

[0101] For example, the credibility of evidence directly determines whether the materials are authentic and valid, and has the greatest impact on risk; therefore, the weight of the inverse value of the credibility of evidence is set at 0.4. The material anomaly directly reflects whether forgery exists; therefore, the weight of the material anomaly is set at 0.3. The information deviation reflects the accuracy of the information; therefore, the weight of the information deviation is set at 0.15. The historical repetition rate reflects the frequency of material use; therefore, the weight of the historical repetition rate is set at 0.15.

[0102] S126. The sum of the products of the information deviation degree and the information deviation degree weight, the product of the evidence credibility reverse value and the evidence credibility reverse value weight, the product of the historical repetition rate and the historical repetition rate weight, and the product of the material anomaly degree and the material anomaly degree weight for each supporting material shall be used as the material risk coefficient for each supporting material.

[0103] In this implementation, the comprehensive risk value of a single supporting material can be calculated by combining the four risk indicators and their corresponding weights that have been extracted, according to the weighted summation rule. The total obtained by weighted summation is used as the material risk coefficient of the supporting material, thereby achieving a multi-dimensional comprehensive assessment of the risk of a single material.

[0104] For example, for a particular piece of supporting material, extract the information deviation degree, information deviation degree weight, evidence credibility reverse value, evidence credibility reverse value weight, historical repetition rate, historical repetition rate weight, material anomaly degree, and material anomaly degree weight, calculate the product of the four dimensions in turn, and then add them together. The sum obtained is the material risk coefficient of the supporting material.

[0105] For example, using the previously extracted project acceptance report indicators and weights: Information deviation is 0.1, and its weight is 0.15. The reverse value of evidence credibility is approximately 1.11, and its weight is 0.4. The historical repetition rate is 0.5, and its weight is 0.15. The material anomaly is 0, and its weight is 0.3. The products are calculated sequentially: 0.1 × 0.15 = 0.015, 1.11 × 0.4 = 0.444, 0.5 × 0.15 = 0.075, and 0 × 0.3 = 0. The sum of these four products is: 0.015 + 0.444 + 0.075 + 0 = 0.534. This sum is the material risk coefficient of this project acceptance report.

[0106] This approach extracts features from four dimensions of supporting materials: information deviation, reverse value of evidence credibility, historical repetition rate, and material anomaly. It can comprehensively cover multiple risk points of supporting materials, such as content deviation, credibility, reuse, and abnormal content, avoiding the problem of missing risks in single-dimensional assessments and ensuring the comprehensiveness of risk assessment for a single document.

[0107] This implementation method matches the information deviation weight, the evidence credibility reverse value weight, the historical repetition rate weight, and the material anomaly weight to four types of features, respectively. The influence ratio of different dimensions can be flexibly adjusted according to actual verification needs, adapting to the risk emphasis requirements of different professional title application scenarios and improving the scenario adaptability of risk assessment. The material risk coefficient is obtained by summing the products of the four types of features and their corresponding weights, which can quantitatively calculate the comprehensive risk value of a single supporting material, avoiding the subjective bias of human judgment and making the risk assessment results of a single material more objective and consistent.

[0108] Figure 9 A flowchart illustrating the fifth automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 9 As shown, in some implementations, in the above-mentioned S120, the adjustment risk coefficient of each supporting material is determined based on the material risk coefficient of each supporting material through the association edge between multiple supporting materials. It also includes S210 to S220, which will be explained in detail below.

[0109] S210. Determine the citation count of each supporting document. Based on the citation count of each supporting document, determine the citation impact coefficient of each supporting document. Obtain the first-level citation impact coefficient for the first-level related supporting documents, the second-level citation impact coefficient for the second-level related supporting documents, and the third-level citation impact coefficient for the third-level related supporting documents for each supporting document.

[0110] Figure 10 A schematic diagram of the workflow for the fifth automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 10 As shown, in this implementation, all the associated edges in the supporting material association network can be traversed to count the total number of times each supporting material is cited by other supporting materials, thus obtaining the number of times each supporting material is cited, providing basic data for subsequently determining the citation impact coefficient.

[0111] For example, the system traverses all the associated edges pointing to each node in the entire network, counts the number of associated edges received by each node, and this number is the number of times the supporting material of the corresponding node is cited, which is stored in the attribute information of the corresponding material.

[0112] In this implementation, the citation count of each supporting material can be combined with the citation impact coefficient of the corresponding supporting material according to the pre-set corresponding rules, so as to reflect the differentiated impact of materials with different citation counts on the target material.

[0113] For example, the system pre-defines the correspondence between the number of citations and the citation impact coefficient. After counting the number of citations, it determines the interval to which the number of citations belongs and matches the corresponding citation impact coefficient.

[0114] For example, the corresponding rule types include interval segmentation rules and linear positive correlation calculation rules. Interval segmentation rules divide the number of citations into different intervals, with each interval corresponding to a fixed citation impact coefficient. Linear positive correlation calculation rules set the citation impact coefficient to be proportional to the number of citations, and calculate it directly using a formula.

[0115] For example, when using the interval segmentation correspondence rule, 0 references correspond to a reference influence coefficient of 1, 1-3 references correspond to a reference influence coefficient of 1.2, 4-6 references correspond to a reference influence coefficient of 1.5, and 7 or more references correspond to a reference influence coefficient of 1.8.

[0116] In this implementation, the reference influence coefficients corresponding to each level of related supporting materials can be extracted by combining the first-level, second-level, and third-level related supporting materials that have already been classified into levels, and the reference influence coefficients of the corresponding levels can be compiled into a set.

[0117] For example, after completing the hierarchical division of the target supporting materials, the citation impact coefficients of all first-level related supporting materials are extracted and organized into first-level citation impact coefficients, the citation impact coefficients of all second-level related supporting materials are extracted and organized into second-level citation impact coefficients, and the citation impact coefficients of all third-level related supporting materials are extracted and organized into third-level citation impact coefficients.

[0118] For example, the primary supporting material for the target paper is a project acceptance report, which has been cited 5 times, resulting in a citation impact coefficient of 1.5. Therefore, the primary citation impact coefficient for the primary supporting material of the target paper is also 1.5. The secondary supporting material for the target paper is a professional qualification certificate, which has been cited 2 times, resulting in a citation impact coefficient of 1.2. Therefore, the secondary citation impact coefficient for the secondary supporting material is also 1.2. The tertiary supporting material for the target paper is an officially issued document, which has been cited 0 times, resulting in a citation impact coefficient of 1. Therefore, the tertiary citation impact coefficient for the tertiary supporting material is also 1, completing the entire acquisition process.

[0119] S220. Determine the product of the first-level association weight, the first-level material risk coefficient, and the first-level citation impact coefficient for each supporting material, as the first risk coefficient. Determine the product of the second-level association weight, the second-level material risk coefficient, and the second-level citation impact coefficient for each supporting material, as the second risk coefficient. Determine the product of the third-level association weight, the third-level material risk coefficient, and the third-level citation impact coefficient for each supporting material, as the third risk coefficient. Determine the sum of the product of the material risk coefficient and the citation impact coefficient, the first risk coefficient, the second risk coefficient, and the third risk coefficient for each supporting material, as the adjusted risk coefficient for each supporting material.

[0120] In this implementation, the first-level association weight, first-level material risk coefficient, and first-level citation influence coefficient of the first-level related supporting materials corresponding to the target supporting materials can be extracted. The three parameters are multiplied continuously, and the product is used as the first risk coefficient corresponding to the target supporting materials, reflecting the comprehensive risk contribution of the first-level related materials.

[0121] For example, for the supporting materials for the target, the corresponding first-level related parameters are retrieved from the stored database, the multiplication of the three parameters is directly performed, and the calculation result is stored as the first risk coefficient.

[0122] For example, the primary supporting material for the target paper is the project acceptance report, with a primary association weight of 0.8, a primary material risk coefficient of 0.534, and a primary citation impact coefficient of 1.5. The calculation process is 0.8 × 0.534 × 1.5 ≈ 0.641, and the resulting product 0.641 is the primary risk coefficient.

[0123] In this implementation, the secondary association weight, secondary material risk coefficient, and secondary citation influence coefficient of the secondary related supporting materials corresponding to the target supporting materials can be extracted. The three parameters are multiplied continuously, and the product is used as the second risk coefficient corresponding to the target supporting materials, reflecting the comprehensive risk contribution of the secondary related materials.

[0124] For example, for the supporting materials for the target, the corresponding secondary correlation parameters are retrieved from the stored database, the multiplication of the three parameters is directly completed, and the calculation result is stored as the second risk coefficient.

[0125] For example, taking the above target paper case as an example, the secondary related supporting material is the professional qualification certificate, the secondary related weight is 0.4, the secondary material risk coefficient is 0.1, the secondary citation impact coefficient is 1.2, the calculation process is 0.4×0.1×1.2=0.048, and the product 0.048 is the second risk coefficient.

[0126] In this implementation, the third-level association weight, third-level material risk coefficient, and third-level citation influence coefficient of the third-level related supporting materials corresponding to the target supporting materials can be extracted. The three parameters are multiplied continuously, and the product is used as the third risk coefficient corresponding to the target supporting materials, reflecting the comprehensive risk contribution of the third-level related materials.

[0127] For example, for the supporting materials for the target, the corresponding third-level correlation parameters are retrieved from the stored database, the multiplication of the three parameters is directly performed, and the calculation result is stored as the third risk coefficient.

[0128] For example, taking the above target paper case as an example, the third-level association supporting material is the officially issued document, the third-level association weight is 0.1, the third-level material risk coefficient is 0.05, and the third-level citation impact coefficient is 1. The calculation process is 0.1×0.05×1=0.005, and the resulting product 0.005 is the third risk coefficient.

[0129] In this implementation method, the product of the material risk coefficient and the citation influence coefficient of the target supporting material itself can be calculated first. Then, the product is added to the first risk coefficient, the second risk coefficient, and the third risk coefficient. The sum obtained is the final adjusted risk coefficient of the target supporting material, which integrates the target's own risk and the comprehensive risk contribution of related materials at different levels.

[0130] For example, extract the material risk coefficient and citation impact coefficient of the supporting material itself, calculate the product of the two, and then add the first risk coefficient, second risk coefficient, and third risk coefficient that have been calculated in sequence. The sum obtained is the adjusted risk coefficient.

[0131] For example, continuing with the target paper case above, the target paper's own material risk coefficient is 0.2, and its own citation impact coefficient is 1.2, with a product of 0.2 × 1.2 = 0.24. Adding the first risk coefficient of 0.641, the second risk coefficient of 0.048, and the third risk coefficient of 0.005, the total is 0.24 + 0.641 + 0.048 + 0.005 = 0.934, which is the adjusted risk coefficient of the target paper.

[0132] This implementation method counts the number of times each supporting material is cited and calculates the citation impact coefficient for each supporting material. This quantifies the scope of the associated impact of supporting materials, fully considers the risk transmission effect of materials with a high number of citations on other related materials, and avoids the risk calculation bias caused by ignoring the citation influence of materials.

[0133] This implementation method combines the first-level, second-level, and third-level citation impact coefficients with the corresponding level's association weight and material risk coefficient to obtain the first risk coefficient, second risk coefficient, and third risk coefficient. This allows the citation influence of materials at different levels to be included in the risk transmission calculation, further improving the accuracy of hierarchical association risk calculation. The adjusted risk coefficient is obtained by summing the product of its own material risk coefficient and its own citation impact coefficient with the three levels of risk coefficients. This can simultaneously cover its own risk, its own citation influence, and the risks and citation influence of multi-level associated materials, making the calculation result of the adjusted risk coefficient more consistent with the actual comprehensive risk level of the material.

[0134] Figure 11 A flowchart illustrating the sixth automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 11 As shown, in some implementations, in S210 above, the citation impact coefficient of each supporting material is determined based on the number of times each supporting material is cited, including S211 to S212. S211 to S212 will be explained in detail below.

[0135] S211. Obtain the supporting material type for each supporting material. Obtain the citation material weight corresponding to the supporting material type of the supporting material.

[0136] Figure 12 A schematic diagram of the workflow for the sixth automated compliance verification method for professional title application materials provided in this application embodiment is shown below. Figure 12 As shown, in this implementation, the type of each uploaded supporting material can be identified to determine the type of supporting material corresponding to each supporting material, providing a classification basis for subsequent matching of the weight of the corresponding cited materials.

[0137] For example, the system can determine the type of supporting material for each document based on the material labels filled in by the applicant, or by recognizing the content of the supporting materials through a content recognition algorithm, and record it in the material attributes.

[0138] For example, common types of supporting materials for professional title applications include basic identity materials, qualification materials, performance achievement materials, academic achievement materials, continuing education materials, and honor and award materials. Each type of material corresponds to different evaluation and assessment directions and plays a different role in the professional title evaluation process.

[0139] For example, basic identity documents include ID cards, household registration books, etc., used to prove the applicant's basic identity. Performance achievement documents include project acceptance reports, award certificates, etc., used to prove the applicant's work performance.

[0140] In this implementation, for the identified supporting material type, the pre-set weight of the cited material for that type can be matched to obtain the weight of the cited material for the corresponding supporting material type, providing weight parameters for subsequent weighted calculation.

[0141] For example, the system pre-stores the correspondence between different types of supporting materials and the weights of cited materials. After identifying the type of supporting material, it directly retrieves the corresponding weight of cited material according to the correspondence.

[0142] For example, the matching rule types for different types of supporting materials and their corresponding reference material weights include fixed-value matching rules and hierarchical mapping matching rules. Fixed-value matching rules assign a fixed reference material weight to each type of supporting material. Hierarchical mapping rules divide material types into different importance levels, with the same weight corresponding to the same level.

[0143] For example, when using the fixed value matching rule, the weight of the cited materials corresponding to performance achievement materials is 1.2, the weight of the academic achievement materials is 1.2, the weight of the basic identity materials is 0.8, and the weight of the continuing education materials is 0.9.

[0144] For example, if a supporting document is identified as a project acceptance report, which is of the performance achievement category, the system retrieves the weight 1.2 of the referenced material corresponding to the performance achievement category in the pre-matching rules, binds this weight to the project acceptance report, and completes the acquisition process.

[0145] For example, if a supporting document is identified as a continuing education completion certificate, and the corresponding type is continuing education material, the system retrieves the reference material weight of 0.9 corresponding to the continuing education category in the pre-matching rules, binds this weight to this completion certificate, and completes the acquisition process.

[0146] S212. Determine the product of the number of citations for each supporting material and the citation weight corresponding to the type of supporting material, as the basic citation impact coefficient. Determine the quotient of the basic citation impact coefficient of each supporting material divided by the number of supporting materials for multiple supporting materials, as the citation impact coefficient of each supporting material.

[0147] In this implementation, the number of citations for each piece of supporting material can be extracted, along with the citation weight corresponding to the type of supporting material. The two parameters are multiplied to obtain a product, which is the basic citation influence coefficient of the corresponding supporting material, integrating the influence of the two dimensions of the number of citations and the material type.

[0148] For example, for a specific piece of supporting material, the number of times it has been cited and the corresponding weight of the cited material can be retrieved from the material attributes, the multiplication can be performed directly, and the calculation result can be stored as the basic citation influence coefficient.

[0149] For example, a project acceptance report is a supporting document for performance results. It has been cited 5 times. The weight of the cited document for performance results is 1.2. The calculation process is 5 × 1.2 = 6. The product 6 is the basic citation impact coefficient of the project acceptance report.

[0150] In this implementation, the total number of all supporting materials uploaded in this application can be counted. The basic citation impact coefficient of each supporting material is divided by the total number, and the quotient is the final citation impact coefficient of the corresponding supporting material. The result is standardized to eliminate the interference of the difference in the total number of application materials on the range of citation impact coefficient values.

[0151] For example, after calculating the total number of supporting materials uploaded for this professional title application, the basic citation impact coefficient of each material is extracted, and the final citation impact coefficient is obtained by performing division.

[0152] For example, in the case of the above-mentioned project acceptance report, a total of 8 supporting materials were uploaded for this professional title application. The basic citation impact coefficient of the project acceptance report is 6. The calculation process is 6÷8=0.75. The quotient 0.75 is the final citation impact coefficient of the project acceptance report.

[0153] This implementation method obtains the supporting material type for each supporting material and matches the corresponding cited material weight. It can distinguish the differences in citation value of different types of supporting materials, avoid the problem of consistent contribution of different types of supporting materials in the calculation of citation impact, and ensure the rationality of citation impact assessment. The basic citation impact coefficient is obtained by multiplying the number of times each supporting material is cited and the corresponding cite material weight. The basic impact level can be quantified by combining the citation scale and material type attributes, fully reflecting the actual citation coverage of the material and improving the accuracy of basic impact assessment.

[0154] By dividing the basic citation impact coefficient by the number of supporting materials to obtain the citation impact coefficient, the interference of the overall scale of the application materials can be eliminated, and the citation impact coefficients under different application batches and different quantities of materials can be horizontally comparable, which is convenient for uniform inclusion in the subsequent risk calculation process.

[0155] In some implementations, in S125 above, the information deviation of each supporting material is obtained, including S125a to S125b. S125a to S125b will be explained in detail below.

[0156] S125a. Extract the key and non-key data fields for each supporting document. Obtain the official key and non-key data fields corresponding to the key data fields of each supporting document retrieved from the official database. The key data fields include name, date, rank, position, and number.

[0157] In this implementation, each uploaded supporting material can be split into content fields. Based on the degree of influence of the fields on the authenticity and validity of the material, key fields that affect the authenticity of the core information and non-key fields that do not affect the authenticity of the core information can be distinguished, providing a hierarchical basis for subsequent calculation of information deviation.

[0158] For example, the system uses OCR to identify and extract all editable text content from supporting materials. Combined with a pre-defined list of core fields, the matched core fields are categorized as key fields of the material, while the remaining content is categorized as non-key fields.

[0159] For example, common types of key fields in supporting materials for professional title applications include name, certificate number, date, project level, job title, certificate number, and approval document number. These fields directly affect the core authenticity of the materials. Common types of non-key fields include information on minor personnel in the project participant list, header and footer information, formatting information, and redundant background descriptions of the project content.

[0160] For example, for a project acceptance report bearing an official seal, the system first uses OCR to extract all text content, matches it against a preset list of core fields, and extracts the applicant's name, project approval date, project level, applicant's position in the project, and project approval number. This information is then categorized into the key fields of the material. The specific description of the project acceptance process, redundant information about the organization name in the header, and information about project participants other than the applicant are categorized into the non-key fields of the material, completing the extraction process.

[0161] In this implementation, the key material fields of each supporting material extracted can be sent to the official database through an interface that connects to the official database. The official key fields and official non-key fields of the corresponding supporting materials stored in the official database can be retrieved to obtain official standard information for deviation comparison.

[0162] For example, the system connects to the official database of the corresponding functional department through an encrypted interface, uses key material fields as query conditions, sends a query request, receives standard information returned by the official database, and separates the official key fields and official non-key fields.

[0163] For example, when querying the official information corresponding to a project acceptance report, the system uses the extracted project approval number as the query condition and sends a query request to the official project database of the science and technology department. The database returns all the filing information for the project. The system separates the officially stored name, date, level, position, and number from the filing information as official key fields, and separates the remaining filing content as official non-key fields, thus completing the query process.

[0164] For example, when querying the official information corresponding to a professional qualification certificate, the system uses the extracted certificate number as the query condition, sends a query request to the official professional qualification database of the human resources and social security department, the database returns all the filing information of the certificate, separates the official key fields and official non-key fields, and completes the query process.

[0165] S125b: When the material's key fields and the official key fields are completely identical, and the material's non-key fields and the official non-key fields are completely identical, the information deviation is determined to be 0. When the material's key fields and the official key fields are completely identical, and the material's non-key fields and the official non-key fields are not completely identical, the information deviation is determined to be 0.2. When the material's key fields and the official key fields are not completely identical, the information deviation is determined to be 0.6. When no official key field corresponding to the material's key field is found in the official database, the information deviation is determined to be 0.8. When the number of inconsistencies between the material's key fields and the official key fields exceeds a preset number of inconsistencies, the information deviation is determined to be 1.

[0166] In this implementation, the extracted key and non-key material fields can be compared with the official key and non-key fields returned by the official database. Based on different comparison results, the corresponding preset information deviation value is matched to obtain the final information deviation of the supporting material.

[0167] For example, the system compares and judges in order of priority from high to low. First, it judges whether there are cases where no results were found. Then, it judges whether the number of inconsistencies exceeds the threshold. Next, it judges whether the key fields are consistent. Finally, it judges whether the non-key fields are consistent, and obtains the corresponding information deviation.

[0168] For example, if an applicant uploads a professional qualification certificate, all key fields such as name, date, level, position, and number are completely consistent with the official key fields, and all non-key fields in the materials are also completely consistent with the official non-key fields, then the application meets the conditions for this judgment scenario, and the corresponding information deviation is set to 0.

[0169] For example, in a project acceptance report uploaded by an applicant, all the key fields in the materials are completely consistent with the official key fields, but the order of project participants in the non-key fields is inconsistent with the official filing information. The other non-key fields are consistent, which meets the judgment scenario conditions, and the corresponding information deviation degree is set to 0.2.

[0170] For example, in a project acceptance report uploaded by an applicant, there is a discrepancy between the project level and the official key field in one material, while the other key fields are consistent. This meets the criteria for determining that the key fields are not completely consistent, and the corresponding information deviation is set to 0.6.

[0171] For example, an applicant uploads a self-compiled set of achievement materials and uses the self-filled project number as the key field of the material to query the official project database. There is no project filing information with the corresponding number in the official database, and no corresponding official key field is found. This meets the judgment scenario conditions, and the corresponding information deviation degree is set to 0.8.

[0172] For example, with a preset inconsistency count of 1, if an applicant uploads a forged honorary certificate, there are four material key fields that are inconsistent with the official key fields: name, date, level, and number. The inconsistency count is 4, which is greater than the preset inconsistency count of 1. This meets the judgment scenario condition, and the corresponding information deviation degree is set to 1.

[0173] This implementation method distinguishes between key and non-key fields in supporting materials, and then compares them with the official key and non-key fields in the official database. This clarifies the differences in field importance, focuses on verifying the accuracy of core information such as name and date, avoids non-core information deviations from interfering with core verification results, and improves the targeting of information deviation identification.

[0174] This implementation method sets a gradient information deviation degree from 0 to 1 according to different field matching situations. This can quantify the degree of information deviation in different matching scenarios, clearly distinguish the risk levels of different problems such as field mismatch, lack of official records, and multiple errors in core fields, avoid the ambiguity of qualitative judgment, and ensure the accuracy of information deviation degree assignment.

[0175] In some implementations, in step S125 above, the reverse credibility value of each supporting document is obtained, including: when the supporting document is verified in an official database, the reverse credibility value is determined to be 0.1; when the supporting document is a scanned copy of the original certificate issued by an official organization, the reverse credibility value is determined to be 0.3; when the supporting document is a certificate issued by an organization and stamped with its official seal, the reverse credibility value is determined to be 0.5; when the supporting document is a document filled out by the applicant without supporting documentation, the reverse credibility value is determined to be 0.8.

[0176] In this implementation, different credibility levels can be distinguished based on the source and verification status of the supporting materials. The corresponding preset evidence credibility reverse value is matched according to the different levels to obtain the final evidence credibility reverse value of the supporting materials, which reflects the credibility level of the supporting materials in reverse.

[0177] For example, the system categorizes supporting materials in descending order of credibility. The higher the credibility, the smaller the corresponding inverse credibility value of the evidence. The system matches the corresponding preset value according to the classification results.

[0178] For example, common scenarios where supporting materials are verified in official databases include: the application project for professional title being verified in the official project database of the science and technology department; the professional qualification certificate being verified in the official professional qualification database of the human resources and social security department; and the continuing education certificate being verified in the official continuing education system of the housing and construction department.

[0179] For example, if an applicant uploads an intermediate professional title qualification certificate, and the information is verified by connecting to the official database of the human resources and social security department, and the information matches, then the scenario is one where the certificate has been verified in the official database, and the corresponding evidence credibility value is 0.1.

[0180] For example, common types of supporting materials such as scanned copies of official certificates include scanned copies of official professional title certificates, scanned copies of official honor and award certificates, scanned copies of official vocational qualification certificates, and scanned copies of official achievement appraisal certificates.

[0181] For example, a scanned copy of the original provincial science and technology progress award certificate uploaded by an applicant is a type of supporting material that is an original certificate issued by the government. The corresponding evidence credibility value is 0.3.

[0182] For example, common types of supporting documents issued by an organization and stamped with its official seal include proof of the applicant's length of service, proof of project participation, proof of workload, and proof of awards received.

[0183] For example, if an applicant uploads a certificate of appointment as the project leader issued by their employer and stamped with the employer's official seal, it is considered a certificate issued by the employer and stamped with the official seal, and the corresponding evidence credibility value is 0.5.

[0184] For example, common types of self-filled lists of achievements, lists of papers, resumes, and continuing education experiences submitted by applicants without supporting documentation are provided by the applicants themselves.

[0185] For example, an applicant filled in a list of five papers without uploading the original papers or retrieval proofs. This is considered unsupported material filled in by the applicant, and the corresponding evidence credibility value is 0.8.

[0186] This implementation method categorizes credibility levels based on the source and verification status of supporting materials, and matches different inverse credibility values ​​accordingly. This accurately reflects the differences in credibility levels among different types of supporting materials, avoids the problem of inconsistent credibility assessments for materials with different probative effects, and ensures the rationality of credibility quantification.

[0187] With this implementation, the reverse credibility value of evidence increases as the evidentiary value decreases, which can be directly connected to the subsequent risk calculation logic without the need for additional conversion of credibility indicators, thus simplifying the risk calculation process and improving the overall verification efficiency.

[0188] In some implementations, in step S125 above, the historical duplication rate and material anomaly degree of each supporting document are obtained, including: retrieving usage records of supporting documents from the historical application database. When a supporting document is used for the first time, the historical duplication rate is determined to be 0. When the same applicant has used supporting documents in historical application records at different levels and it has been noted as repeated use, the historical duplication rate is determined to be 0.2. When a supporting document has been used by other applicants, the historical duplication rate is determined to be 0.7. When a supporting document is used multiple times in this application and is repeatedly scored, the historical duplication rate is determined to be 0.9.

[0189] In this implementation, each supporting document submitted at present can be submitted to the historical application database for content feature matching and retrieval to obtain the historical usage record of the supporting document. Then, based on different search results, the corresponding preset historical duplication rate is matched to obtain the final historical duplication rate of the supporting document.

[0190] For example, the system extracts the content feature hash value of the supporting materials, compares the hash value with the hash values ​​of all supporting materials stored in the historical declaration database, and if a match is found, the corresponding usage record is extracted; otherwise, it is determined that there is no usage record.

[0191] For example, after extracting the content feature hash value of the current supporting materials, the hash value is sent to the historical declaration database. The database compares the hash values ​​of all stored historical materials and returns all matching usage record information, including the user, declaration level, and number of uses, thus completing the retrieval process.

[0192] For example, if an applicant is applying for a senior professional title for the first time, and the project acceptance report they uploaded has no record of being used in the historical application database, they meet the criteria for first-time use, and the corresponding historical duplication rate is 0.

[0193] For example, if an applicant has already passed the intermediate professional title review and is now applying for a senior professional title, the project acceptance report used in the previous intermediate application has been noted as a duplicate material in this application, which meets the criteria for this judgment scenario, and the corresponding historical duplication rate is 0.2%.

[0194] For example, if an applicant uploads a paper and it is found in the historical application database that the paper has been used by another applicant for a senior professional title application in the previous year, it meets the criteria for being used by another applicant, and the corresponding historical duplication rate is 0.7.

[0195] For example, an applicant applying for a senior professional title used the same project acceptance report to be scored in both the personal performance achievement and academic achievement modules. This meets the criteria for multiple uses and repeated scoring in this application, with a corresponding historical duplication rate of 0.9.

[0196] In some implementations, obtaining the historical duplication rate and material anomaly degree for each supporting material in S125 above further includes: identifying timeline crossover anomalies, numerical anomalies, and homonymy anomalies in the supporting materials. When supporting materials exhibit timeline crossover anomalies, numerical anomalies, and homonymy anomalies, the material anomaly degree is determined to be 1. When supporting materials exhibit only timeline crossover anomalies and numerical anomalies, the material anomaly degree is determined to be 0.8. When supporting materials exhibit only timeline crossover anomalies, the material anomaly degree is determined to be 0.3. When supporting materials exhibit no anomalies, the material anomaly degree is determined to be 0. Timeline crossover anomalies include results dates earlier than the start date; numerical anomalies include discrepancies between the actual journal quartile and the journal's labeled quartile, and project funding exceeding or equaling the regular project funding of similar supporting materials; homonymy anomalies include supporting materials with authors sharing the same name but different affiliations.

[0197] In this implementation, each piece of supporting material can be checked for three types of anomalies: timeline crossing anomaly, numerical anomaly, and homonymy anomaly. The number of anomaly types is counted, and the corresponding preset material anomaly degree is matched according to the anomaly combination to obtain the final material anomaly degree of the supporting material.

[0198] For example, the system sequentially completes the detection of three types of anomalies, records the detection results, and matches them with the corresponding preset material anomaly values.

[0199] For example, common scenarios involving timeline anomalies include an applicant joining the company in 2015, with the submitted deliverables completed in 2012; or an applicant obtaining an intermediate professional title in 2018, with the submitted project deliverables completed in 2016.

[0200] For example, common scenarios of numerical anomalies include: the applicant marking the journal to which they submitted as Q1, but the journal was actually in Q3 that year; the applicant requesting 50 million in funding, while the usual funding for similar projects at the same level does not exceed 10 million; the applicant marking the impact factor as 5.2, but the actual impact factor of the paper is 2.1.

[0201] For example, common scenarios of name confusion include applicants named Zhang San and authors of the journal to which they submit their papers, but the authors' affiliations are different universities, not the applicant's. Another example is an applicant named Li Ming, whose award-winning work is authored by an institution other than the applicant's.

[0202] For example, an applicant's start date is 2015, but the submitted research results are dated 2012, indicating a timeline crossing anomaly. The submitted paper is labeled as Q1 but is actually Q3, indicating a numerical anomaly. The submitted paper has authors with the same name but different affiliations, indicating a name confusion anomaly. With all three anomalies present, the corresponding material anomaly score is 1.

[0203] For example, an applicant's start date is 2015, but the date of the submitted achievement is 2012, which constitutes a timeline anomaly. The submitted project funding exceeds the regular funding for similar projects, which is a numerical anomaly. There are no anomalies related to name confusion; only these two anomalies exist, corresponding to an anomaly score of 0.8.

[0204] For example, an applicant joined the company in 2015, but the date of the submitted achievement was 2012, which constitutes a timeline crossing anomaly. There are no numerical anomalies or name confusion anomalies; only one anomaly exists, corresponding to an anomaly degree of 0.3.

[0205] For example, in all the supporting materials of an applicant, there are no timeline anomalies where the achievement date is later than the start date. The paper's section markings are consistent with the actual situation, and the project funding is in accordance with the norm with no numerical anomalies. All paper authors' affiliations and applicants are consistent, so there are no anomalies due to name confusion. There are no anomalies at all, and the corresponding material anomaly degree is 0.

[0206] This implementation method retrieves the usage records of supporting materials from the historical application database and assigns historical duplication rates according to different reuse scenarios. This can accurately distinguish the risk levels of different reuse situations, avoid misjudging normally reused materials, and identify illegal reuse issues such as theft and duplicate scoring, thereby improving the accuracy of historical reuse risk assessment.

[0207] This implementation method uses timeline anomalies, numerical anomalies, and name confusion anomalies in supporting materials as the criteria for determining the degree of anomaly in the materials. It covers three common anomaly scenarios: logical content of materials, data authenticity, and subject matching, avoiding omission of hidden content issues and ensuring the comprehensiveness of anomaly identification.

[0208] This application also provides an automated compliance verification system for professional title application materials, including units for implementing the method described above.

[0209] Figure 13 A schematic diagram of an automated compliance verification system for professional title application materials provided in this application embodiment is shown below. Figure 13As shown, the system 1 of this embodiment includes a processing unit 11, a storage unit 12, and a transceiver unit 13. The processing unit 11 is used to process data, the storage unit 12 is used to store data, and the transceiver unit 13 is used to send and receive data. The processing unit 11, the storage unit 12, and the transceiver unit 13 cooperate with each other to implement the above-described method. The beneficial effects of the embodiments of this application have been described in the above-described method and will not be repeated here.

[0210] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0211] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0212] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0213] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0214] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0215] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0216] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0217] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An automated compliance verification method for professional title application materials, characterized in that, The method includes: Obtain the citation relationships between multiple supporting materials used for professional title application; construct a supporting material association network by treating each supporting material as a node and the citation relationships between multiple supporting materials as association edges; Obtain the material risk coefficient for each supporting document; determine the adjustment risk coefficient for each supporting document based on the material risk coefficient of each supporting document through the association edges between multiple supporting documents; determine the application risk coefficient corresponding to multiple supporting documents based on the adjustment risk coefficient of multiple supporting documents; mark multiple supporting documents as unqualified when the application risk coefficient is greater than or equal to the preset application risk coefficient; mark multiple supporting documents as qualified when the application risk coefficient is less than the preset application risk coefficient.

2. The method according to claim 1, characterized in that, By identifying the correlation edges between multiple supporting documents, and based on the material risk coefficient of each supporting document, the adjustment risk coefficient for each supporting document is determined, including: By identifying the association edges between multiple supporting materials, we determine the first-level associated supporting materials, the second-level associated supporting materials, and the third-level associated supporting materials, which are associated through adjacent nodes and one intermediate node, respectively. We then obtain the first-level material risk coefficient, the second-level material risk coefficient, and the third-level material risk coefficient for each supporting material. Obtain the first-level association weights corresponding to the first-level association supporting materials, the second-level association weights corresponding to the second-level association supporting materials, and the third-level association weights corresponding to the third-level association supporting materials; determine the material risk coefficient of each supporting material, the product of the first-level association weight and the first-level material risk coefficient, the product of the second-level association weight and the second-level material risk coefficient, and the product of the third-level association weight and the third-level material risk coefficient, as the adjusted risk coefficient of each supporting material.

3. The method according to claim 2, characterized in that, Based on the adjustment risk coefficients of multiple supporting documents, the corresponding application risk coefficients for multiple supporting documents are determined, including: Obtain the weight of each supporting document; The sum of the products of the adjustment risk coefficients of multiple supporting materials and the weights of the supporting materials is determined as the aggregated declaration risk coefficient; the quotient of the aggregated declaration risk coefficient divided by the number of supporting materials for the multiple supporting materials is determined as the declaration risk coefficient corresponding to the multiple supporting materials.

4. The method according to claim 3, characterized in that, Obtain the material risk coefficient for each supporting document, including: The information deviation degree, evidence credibility reverse value, historical duplication rate, and material anomaly degree of each supporting material are obtained; the information deviation degree weight, evidence credibility reverse value weight, historical duplication rate weight, and material anomaly degree weight are obtained; among them, the information deviation degree represents the degree of information deviation of the supporting materials, the evidence credibility reverse value is the reciprocal of the credibility of the supporting materials as supporting evidence, the historical duplication rate represents the duplication rate of the supporting materials as professional title application materials within a preset time period, and the material anomaly degree represents the degree of content anomaly of the supporting materials; The sum of the products of information deviation degree and information deviation degree weight, evidence credibility reverse value and evidence credibility reverse value weight, historical repetition rate and historical repetition rate weight, and material anomaly degree and material anomaly degree weight for each supporting material is determined as the material risk coefficient for each supporting material.

5. The method according to claim 4, characterized in that, By identifying the correlation edges between multiple supporting materials, and based on the material risk coefficient of each supporting material, the adjustment risk coefficient for each supporting material is determined. This also includes: Determine the number of times each supporting material is cited; based on the number of times each supporting material is cited, determine the citation impact coefficient of each supporting material; obtain the first-level citation impact coefficient corresponding to the first-level related supporting materials, the second-level citation impact coefficient corresponding to the second-level related supporting materials, and the third-level citation impact coefficient corresponding to the third-level related supporting materials for each supporting material; The product of the first-level association weight, the first-level material risk coefficient, and the first-level citation impact coefficient for each supporting material is determined as the first risk coefficient; the product of the second-level association weight, the second-level material risk coefficient, and the second-level citation impact coefficient for each supporting material is determined as the second risk coefficient; the product of the third-level association weight, the third-level material risk coefficient, and the third-level citation impact coefficient for each supporting material is determined as the third risk coefficient; and the sum of the product of the material risk coefficient and the citation impact coefficient for each supporting material, the first risk coefficient, the second risk coefficient, and the third risk coefficient is determined as the adjusted risk coefficient for each supporting material.

6. The method according to claim 5, characterized in that, Based on the number of times each supporting material is cited, determine the citation impact coefficient of each supporting material, including: Obtain the supporting material type for each supporting material; obtain the citation material weight corresponding to the supporting material type of each supporting material. The product of the number of citations for each supporting material and the weight of the citation material corresponding to the type of supporting material is determined as the basic citation impact coefficient; the quotient of the basic citation impact coefficient of each supporting material divided by the number of supporting materials for multiple supporting materials is determined as the citation impact coefficient of each supporting material.

7. The method according to claim 6, characterized in that, Obtain the information bias of each supporting document, including: Extract the key and non-key fields of each supporting document; obtain the official key and non-key fields corresponding to the key fields of each supporting document retrieved from the official database; among them, the key fields of the supporting document include name, date, level, position and number; When the material's key fields and the official key fields are completely identical, and the material's non-key fields and the official non-key fields are completely identical, the information deviation is determined to be 0. When the material's key fields and the official key fields are completely identical, and the material's non-key fields and the official non-key fields are not completely identical, the information deviation is determined to be 0.

2. When the material's key fields and the official key fields are not completely identical, the information deviation is determined to be 0.

6. When no supporting material key fields corresponding to the official key fields are found in the official database, the information deviation is determined to be 0.

8. When the number of inconsistencies between the material's key fields and the official key fields exceeds the preset number of inconsistencies, the information deviation is determined to be 1.

8. The method according to claim 7, characterized in that, Obtain the inverse credibility value for each piece of supporting evidence, including: When supporting materials are verified in the official database, the reverse credibility value is determined to be 0.1; when the supporting materials are scanned copies of the original certificates issued by the authorities, the reverse credibility value is determined to be 0.3; when the supporting materials are certificates issued by the organization and stamped with the official seal, the reverse credibility value is determined to be 0.5; when the supporting materials are self-filled by the applicant without supporting materials, the reverse credibility value is determined to be 0.

8.

9. The method according to claim 8, characterized in that, Obtain the historical repetition rate and material anomalousness of each supporting material, including: The historical application database is searched for records of the use of supporting materials. When the supporting materials are used for the first time, the historical duplication rate is determined to be 0. When the same applicant has used supporting materials in historical application records at different levels and it has been noted that they were used repeatedly, the historical duplication rate is determined to be 0.

2. When the supporting materials have been used by other applicants, the historical duplication rate is determined to be 0.

7. When the supporting materials are used multiple times in this application and are scored repeatedly, the historical duplication rate is determined to be 0.

9. The supporting materials are identified for timeline anomalies, numerical anomalies, and homonym confusion anomalies. When supporting materials exhibit these anomalies, the anomaly score is set to 1. When supporting materials exhibit only timeline and numerical anomalies, the anomaly score is set to 0.

8. When supporting materials exhibit only timeline anomalies, the anomaly score is set to 0.

3. When supporting materials exhibit no anomalies, the anomaly score is set to 0. Timeline anomalies include results dates earlier than the start date; numerical anomalies include discrepancies between the actual journal quartile and the journal's labeled quartile, and project funding exceeding or equal to the standard project funding of similar supporting materials; homonym confusion anomalies include papers with the same author name but different affiliations.

10. An automated compliance verification system for professional title application materials, characterized in that, Includes units for implementing the method of any one of claims 1 to 9.

Images