Data verification method and device, computer equipment and computer readable storage medium

Abstract

The invention provides a data verification method and apparatus, a computer device and a computer readable storage medium. The method comprises the steps of obtaining product basic information and product risk associated information of a detected subject; determining a target verification rule associated with the detected subject based on the product basic information of the detected subject and a preset verification rule database; determining whether the product risk associated information of the detected subject is matched with the target verification rule; and under the condition that the product risk associated information of the detected main body is matched with the target verification rule, determining that the detected main body is a compliant main body, otherwise, determining that the detected main body is a non-compliant main body. According to the technical scheme of the invention, the convenience and accuracy of the property insurance business compliance inspection are improved.

Description

[Technical field]

[0001] The present application relates to the field of data verification technology, and in particular to a data verification method and apparatus, a computer device, and a computer-readable storage medium. [Background technology]

[0002] With the development of society, the awareness of wealth protection of individual and corporate users has gradually increased, and a large number of businesses that provide users with property protection certificates have emerged. In order to ensure the legality of such businesses, relevant national regulatory authorities have set up a variety of laws, regulations, and other restrictive rules for such businesses. At present, the compliance inspection of such businesses mainly adopts manual review. Faced with businesses with different protection contents, manual review obviously has the defects of being difficult, time-consuming, and having a high error rate. At the same time, due to the diversity of the number and types of such businesses, manual review is limited by the business level of the reviewer, and it is impossible to standardize and uniformly handle businesses of different types and contents. The above problems will lead to difficulties in supervising the business of providing users with property protection certificates and complex supervision processes. The difficulties in supervision and the complexity of the supervision process directly affect the management difficulty and service effect of the business issuer for the business.

[0003] Therefore, how to strike a balance between the regulatory rationality and convenience of providing users with property protection certificates has become a technical problem that needs to be solved urgently. [Summary of the invention]

[0004] The embodiments of the present application provide a data verification method and apparatus, a computer device, and a computer-readable storage medium, which aim to solve the technical problem in the related art that a single regulatory means for a business that provides property protection certificates to users is difficult to meet the regulatory convenience requirements and regulatory rationality requirements of the business.

[0005] In a first aspect, an embodiment of the present application provides a data verification method, including:

[0006] Obtaining basic product information and product risk-related information of the subject being tested, wherein the subject being tested is a business that provides property protection certificates to users, the basic product information includes the product category, sales channel, development method, operating area and insurance liability type of the business, and the product risk-related information is information that has an impact on the property protection capability of the business;

[0007] Based on the basic product information of the detected subject and a preset verification rule database, determining a target verification rule associated with the detected subject;

[0008] Determining whether the product risk association information of the detected subject matches the target verification rule;

[0009] In the case where the product risk association information of the detected subject matches the target verification rule, the detected subject is determined to be a compliant subject; otherwise, the detected subject is determined to be a non-compliant subject.

[0010] In one embodiment of the present application, optionally, before acquiring the basic product information and product risk association information of the detected subject, the method further includes:

[0011] Based on the single initial rule information and the verification rule identification model in the initial rule information sample set, the corresponding verification rule of the single initial rule information is determined, and the corresponding verification rule is added to the verification rule database, wherein the verification rule identification model is used to reflect the association between the single initial rule information and the corresponding verification rule in the historical valid rule verification task.

[0012] In one embodiment of the present application, optionally, determining the target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database includes:

[0013] Obtaining an entity set of the preset verification rule database;

[0014] Based on the basic product information of the detected subject, determining a target entity associated with the detected subject in the entity set;

[0015] The verification rule having the target entity is screened in the preset verification rule database as the target verification rule associated with the detected subject.

[0016] In one embodiment of the present application, optionally, determining the target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database includes:

[0017] Determine search keywords based on the basic product information of the detected subject;

[0018] Determining the effective rules for the search keyword hits in the preset verification rule database;

[0019] Based on the effective rules and a preset large language model, determining entities and inter-entity relationships in the effective rules, wherein the preset large language model is used to reflect the association relationship between the description text of the effective rules and the entities and inter-entity relationships in the effective rules;

[0020] The target verification rules are generated based on the entities and the relationships between entities in the effective rules.

[0021] In one embodiment of the present application, optionally, the preset verification rule database includes completeness rules, compliance rules and early warning rules.

[0022] Before obtaining the basic product information and product risk association information of the subject to be tested, the method further includes:

[0023] Acquire the verification requirements of the detected subject, wherein the verification requirements are used to reflect the type of verification that the detected subject needs to undergo, and the verification type includes one or more of completeness verification, compliance verification, and early warning verification;

[0024] The determining of the target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database includes:

[0025] A target verification rule that is associated with the detected subject and matches the verification requirement is selected from the preset verification rule database.

[0026] In one embodiment of the present application, optionally, the determining whether the product risk association information of the detected subject matches the target verification rule includes at least one of the following detection methods:

[0027] Determining whether the value in the product risk association information matches the value range restriction information in the target verification rule;

[0028] Determine whether the segmented words in the product risk association information hit all the keywords in the target verification rules;

[0029] Detecting whether the product risk association information contains designated signature information by means of OCR recognition;

[0030] Check whether the product risk association information includes a specified file.

[0031] In one embodiment of the present application, optionally, the target verification rule includes that if the comprehensive risk score of the product of the detected subject is less than or equal to a predetermined threshold, then

[0032] The determining whether the product risk association information of the detected subject matches the target verification rule includes:

[0033] Determine a first risk parameter corresponding to each item of product risk association information of the detected subject;

[0034] Based on the first risk parameter, determine a second risk parameter corresponding to each plurality of product risk association information of the detected subject;

[0035] Based on the risk parameter sequence obtained by arranging the first risk parameter and the second risk parameter from large to small, a risk degree matrix is ​​constructed, wherein the risk degree matrix is ​​used to reflect the risk level distribution of the detected subject under the individual influence and interactive influence of each product risk association information;

[0036] Determining a comprehensive risk score of the product of the inspected subject based on the risk matrix and the risk parameter sequence;

[0037] If the comprehensive risk score of the product is less than or equal to the predetermined threshold, it is determined that the product risk association information of the detected subject matches the target verification rule; otherwise, it is determined that the product risk association information of the detected subject does not match the target verification rule.

[0038] In one embodiment of the present application, optionally, constructing a risk matrix based on a risk parameter sequence obtained by arranging the first risk parameter and the second risk parameter from large to small includes:

[0039] The element of the first designated row and the first designated column in the risk matrix is ​​set to be the product of the risk parameter of the first designated order and the risk parameter of the second designated order in the risk parameter sequence.

[0040] In one embodiment of the present application, optionally, determining the comprehensive risk score of the product of the detected subject based on the risk matrix and the risk parameter sequence includes:

[0041] Obtaining the rank of the risk matrix as a third risk parameter;

[0042] Obtaining the variance of the risk parameter sequence as a fourth risk parameter;

[0043] The third risk parameter and the fourth risk parameter are normalized, and the results of the normalization are averaged, and the average is used as the comprehensive risk score of the product of the detected subject.

[0044] In a second aspect, an embodiment of the present application provides a data verification device, characterized in that it includes:

[0045] An information acquisition unit, used to acquire basic product information and product risk-related information of a detected subject, wherein the detected subject is a business that provides property protection certificates to users, the basic product information includes the product category, sales channel, development method, operating area and insurance liability type of the business, and the product risk-related information is information that has an impact on the property protection capability of the business;

[0046] A verification rule determination unit, configured to determine a target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database;

[0047] A verification rule matching unit, used to determine whether the product risk association information of the detected subject matches the target verification rule;

[0048] The compliance subject determination unit is used to determine that the detected subject is a compliant subject when the product risk association information of the detected subject matches the target verification rule, otherwise, determine that the detected subject is a non-compliant subject.

[0049] In one embodiment of the present application, optionally, the device further includes:

[0050] A verification rule database setting unit is used to determine the corresponding verification rule of the single initial rule information based on the single initial rule information and the verification rule identification model in the initial rule information sample set before obtaining the basic product information and product risk association information of the inspected subject, and add the corresponding verification rule to the verification rule database, wherein the verification rule identification model is used to reflect the association relationship between the single initial rule information and the corresponding verification rule in the historical valid rule verification task.

[0051] In one embodiment of the present application, optionally, the verification rule determination unit includes:

[0052] The first execution unit is used to obtain an entity set of the preset verification rule database; based on the basic product information of the detected subject, determine a target entity associated with the detected subject in the entity set; and filter the verification rules with the target entity in the preset verification rule database as the target verification rules associated with the detected subject.

[0053] In one embodiment of the present application, optionally, the verification rule determination unit includes:

[0054] The second execution unit is used to determine the search keywords based on the basic product information of the detected subject; determine the effective rules hit by the search keywords in the preset verification rule database; determine the entities and entity relationships in the effective rules based on the effective rules and a preset large language model, wherein the preset large language model is used to reflect the association relationship between the description text of the effective rules and the entities and entity relationships in the effective rules; generate the target verification rules based on the entities and entity relationships in the effective rules.

[0055] In one embodiment of the present application, optionally, the preset verification rule database includes completeness rules, compliance rules and early warning rules, and the device further includes:

[0056] A verification requirement acquisition unit, used to acquire the verification requirement of the detected subject, wherein the verification requirement is used to reflect the type of verification that the detected subject needs to undergo, and the verification type includes one or more of completeness verification, compliance verification, and early warning verification;

[0057] The verification rule determination unit is used to select a target verification rule from the preset verification rule database that is associated with the detected subject and is compatible with the verification requirement.

[0058] In one embodiment of the present application, optionally, the verification rule matching unit is used to perform at least one of the following detection methods:

[0059] Determine whether the numerical value in the product risk association information matches the numerical range restriction information in the target verification rule; determine whether the segmented words in the product risk association information hit all the keywords in the target verification rule; detect whether the product risk association information contains specified signature information through OCR recognition; detect whether the product risk association information includes a specified file.

[0060] In one embodiment of the present application, optionally, the target verification rule includes that the comprehensive risk score of the product of the detected subject is less than or equal to a predetermined threshold, and the verification rule matching unit includes:

[0061] A first risk parameter determination unit, used to determine a first risk parameter corresponding to each item of product risk association information of the detected subject;

[0062] A second risk parameter determination unit, configured to determine, based on the first risk parameter, a second risk parameter corresponding to each of the plurality of product risk association information of the detected subject;

[0063] a matrix construction unit, configured to construct a risk degree matrix based on a risk parameter sequence obtained by arranging the first risk parameter and the second risk parameter from large to small, wherein the risk degree matrix is ​​used to reflect the risk level distribution of the detected subject under the individual influence and interactive influence of each product risk association information;

[0064] A risk score calculation unit, used to determine the comprehensive risk score of the product of the detected subject based on the risk matrix and the risk parameter sequence;

[0065] The third execution unit is used to determine that the product risk association information of the detected subject matches the target verification rule if the comprehensive risk score of the product is less than or equal to the predetermined threshold; otherwise, determine that the product risk association information of the detected subject does not match the target verification rule.

[0066] In one embodiment of the present application, optionally, the matrix construction unit is used to: set the elements of the first specified row and the first specified column in the risk matrix to be the product of the risk parameters of the first specified order and the risk parameters of the second specified order in the risk parameter sequence.

[0067] In one embodiment of the present application, optionally, the risk score calculation unit is used to: obtain the rank of the risk matrix as a third risk parameter; obtain the variance of the risk parameter sequence as a fourth risk parameter; normalize the third risk parameter and the fourth risk parameter, and average the results of the normalization, and use the average as the comprehensive risk score of the product of the inspected subject.

[0068] In a third aspect, an embodiment of the present application provides a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are configured to execute the method described in the first aspect above.

[0069] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the method described in the first aspect above.

[0070] The above technical solution aims at the technical problem that the single regulatory means for the business of providing property protection certificates to users in the relevant technology is difficult to meet the regulatory convenience and rationality requirements of the business. For the business of providing property protection certificates to users, the applicable target verification rules can be automatically screened based on the basic information of its products, and then the product risk-related information can be automatically checked to see if it matches the target verification rules, thereby realizing the automatic judgment of whether the business is compliant. Through this technical solution, the labor cost and time cost of compliance inspection of property insurance business are reduced, and the convenience and accuracy of compliance inspection of property insurance business are improved. While reducing the management cost of the business management party, it can also effectively enhance the effectiveness of compliance review of property insurance business, indirectly improve the experience of insured users, and more safely and effectively protect their property rights and interests.

Brief Description of the Drawings

[0071] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0072] Figure 1 A flow chart of a data verification method according to an embodiment of the present application is shown;

[0073] Figure 2 A block diagram of a computer device according to an embodiment of the present application is shown;

[0074] Figure 3 A block diagram of a computer device according to another embodiment of the present application is shown. [Specific implementation method]

[0075] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0076] Figure 1 A flow chart of a data verification method according to an embodiment of the present application is shown.

[0077] like Figure 1 As shown, a data verification method according to an embodiment of the present application includes:

[0078] Step 102, obtaining basic product information and product risk association information of the subject to be tested.

[0079] Among them, the detected entity is a business that provides property protection certificates to users, including but not limited to property insurance for various property guarantee needs such as property loss insurance, credit guarantee insurance, and liability insurance.

[0080] The basic product information describes the basic information of the business that provides property protection certificates to users, including the product category, sales channel, development method, operating area and insurance liability type of the business. For example, the product category of property insurance is transportation vehicle insurance under property damage insurance, the sales channel is online direct sales and online intermediaries, the development method is independent development by the insurance provider, the business scope is Shanghai, and the insurance liability types include transportation vehicle safety services and renewal liability.

[0081] The product risk-related information is information that has an impact on the property protection capability of the business, or in other words, information that may reduce the risk of the property protection capability provided by the business. Specifically, the product risk-related information includes but is not limited to the proportion of the main service in the insurance to the net premium, the waiting period, the lowest comprehensive solvency adequacy ratio in the last four quarters, the lowest core solvency adequacy ratio in the last four quarters, the lowest liability reserve coverage ratio in the last four quarters, the lowest comprehensive risk level in the last four quarters, and the corporate governance assessment level of the insurance provider.

[0082] Step 104: Determine a target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database.

[0083] The basic product information describes the basic information of the business that provides property protection certificates to users. These basic information can reflect the degree of standardization of the business. Therefore, it can be used as the basic condition for screening the specifications that the business should comply with. Furthermore, the specifications that the business should comply with are the target verification rules associated with the detected subject. The preset verification rule database includes any form of normative documents such as laws, regulations, and ordinances issued by administrative entities at all levels, or verification rules obtained by preliminary screening in any form of normative documents such as laws, regulations, and ordinances issued by administrative entities at all levels, covering all rules that should be followed by businesses that provide property protection certificates to users. On this basis, the target verification rules that are associated with the basic product information of the detected subject can be searched in the library as the specifications that the detected subject should comply with and need to be verified.

[0084] In one possible design, step 104 includes: obtaining an entity set of the preset verification rule database; determining a target entity associated with the detected subject in the entity set based on the basic product information of the detected subject; and screening the verification rules with the target entity in the preset verification rule database as the target verification rules associated with the detected subject.

[0085] An entity refers to the text body used to describe information of any dimension, such as a name, action, organization, expression, quantity, etc. The following is an example of extracting entities from the preset verification rule database.

[0086] For example, the preset verification rule database includes the "Property Insurance Company Insurance Clauses and Insurance Premium Rates Management Measures" promulgated by the China Banking and Insurance Regulatory Commission Order No. 10 of 2021. Article 16 of this document states that property insurance companies shall submit the following materials when submitting insurance clauses and insurance premiums for approval or filing: (ii) Insurance clauses and insurance premium rate texts. Among them, the name entities that can be extracted are "insurance clauses" and "insurance premium rate texts". Furthermore, if the verification of the subject being tested is because the property insurance company submits insurance clauses and insurance premiums for approval or filing for the subject being tested, then the insurance clauses and insurance premium rate texts are the target entities associated with the subject being tested. At this point, the verification rule describing the insurance clauses and insurance premium rate texts, that is, "Property insurance companies shall submit the following materials when submitting insurance clauses and insurance premiums for approval or filing: (ii) Insurance clauses and insurance premium rate texts" can be set as the target verification rule associated with the subject being tested.

[0087] In another possible design, step 104 includes: determining search keywords based on the basic product information of the detected subject; determining valid rules hit by the search keywords in the preset verification rule database; determining entities and entity relationships in the valid rules based on the valid rules and a preset large language model, wherein the preset large language model is used to reflect the association between the description text of the valid rules and the entities and entity relationships in the valid rules; and generating the target verification rules based on the entities and entity relationships in the valid rules.

[0088] Specifically, corresponding search keywords can be preset for different types of businesses. These search keywords include but are not limited to the description of the approval content of any approval link of the business. Hitting valid rules in the preset verification rule database through search keywords is a preliminary screening of the rules. Then, the preliminary screened valid rules are directly input into the big language model, and the big language model automatically outputs the entities and relationships between entities in the valid rules as the basis for the secondary screening of rules. Among them, the big language model includes any model that can understand and generate human language and can be trained in open source.

[0089] For example, after hitting the effective rule "Property insurance companies submitting insurance terms and insurance premiums for approval or filing shall submit the following materials: (ii) Insurance terms and insurance premium texts" in the preset verification rule database, the effective rule and the automatically set prefix "Extract entities and entity relationships within the rule" can be directly input into the large language model, so that the large language model outputs the name entities "Property Insurance Company", "Insurance Terms" and "Insurance Premium Text", as well as the behavior entity "Submitted", and the entity relationship between them is that the property insurance company has submitted the insurance terms and insurance premium texts. At this point, based on these entities and entity relationships, it can be determined that the target verification rule corresponding to the effective rule is "Insurance Terms and Insurance Premium Texts Have Been Submitted".

[0090] In another possible design, before executing the steps of the technical solution, it is necessary to first construct a verification rule database. At this time, the specific method of constructing the verification rule database is: based on the single initial rule information and the verification rule recognition model in the initial rule information sample set, determine the corresponding verification rule of the single initial rule information, and add the corresponding verification rule to the verification rule database.

[0091] Specifically, among the historical tasks, historical valid rule verification tasks whose verification results are compliance, completeness, and controllable risks are selected as samples for training the verification rule recognition model, wherein the input samples used for training the verification rule recognition model are the single initial rule information involved in the historical valid rule verification tasks, and the output samples used for training the verification rule recognition model are the corresponding verification rules associated with the single initial rule information involved in the historical valid rule verification tasks. In this way, the verification rule recognition model actually reflects the association between the single initial rule information and the corresponding verification rules in the historical valid rule verification tasks. Therefore, for each single initial rule information in the initial rule information sample set, its corresponding verification rule can be obtained through the verification rule recognition model, and then the corresponding verification rules of all the single initial rule information in the initial rule information sample set are added to the verification rule database to complete the preset of the verification rule database.

[0092] In this case, the content stored in the preset verification rule database is a more accurate verification rule obtained after preliminary identification of any form of normative documents such as laws, regulations, and rules issued by administrative entities at all levels. In this way, when the verification rules are matched later, the redundant information in the verification rules is reduced, which helps to obtain more accurate verification results more quickly.

[0093] Step 106, determining whether the product risk association information of the detected subject matches the target verification rule.

[0094] Optionally, when using the target verification rules, the entities within the target verification rules can be encoded according to a preset coding method. For example, the insurance terms and insurance premium texts correspond to the preset codes a012 and a013 respectively, and the action of submitting corresponds to the preset code b01. Therefore, all target verification rules can be converted into a coded form in combination with the preset coding table. In the subsequent rule verification process, it can be checked whether the coding obtained by converting the product risk association information of the subject under inspection according to the preset coding table is consistent with the coding of the corresponding target verification rule. If they are consistent, it means that it has passed the rule verification; otherwise, it means that it has not passed the rule verification. Thus, the convenience of rule verification is increased.

[0095] In one possible design, step 106 includes at least one of the following detection methods.

[0096] First, it can be determined whether the value in the product risk association information matches the value range restriction information in the target verification rule. If the value is within the value range indicated by the value range restriction information in the target verification rule, it is determined that the product risk association information of the detected subject matches the target verification rule. Otherwise, it is determined that the product risk association information of the detected subject does not match the target verification rule.

[0097] Secondly, it can be determined whether the segmented words in the product risk association information hit all the keywords in the target verification rule. If so, it means that the product risk association information includes the main content of the target verification rule. At this time, it can be determined that the product risk association information of the detected subject matches the target verification rule. Otherwise, it is determined that the product risk association information of the detected subject does not match the target verification rule.

[0098] Third, OCR recognition can be used to detect whether the product risk-related information contains designated signature information. For businesses that provide property protection certificates to users, the product risk-related information must be endorsed by a designated manager, and the designated signature information is the signature of the designated manager or a signature indicating approval. Therefore, whether the product risk-related information is compliant can be determined by identifying whether the product risk-related information contains designated signature information. At this point, if it is detected that the product risk-related information contains designated signature information, it means that the product risk-related information of the detected subject matches the target verification rules. Otherwise, it is determined that the product risk-related information of the detected subject does not match the target verification rules.

[0099] Fourth, detect whether the product risk association information includes the designated file. This detection method is intended to determine whether the product risk association information of the detected subject is complete, that is, whether the various documents required for verification have been submitted. If the product risk association information includes the designated file, it means that the product risk association information of the detected subject matches the target verification rule. Otherwise, it is determined that the product risk association information of the detected subject does not match the target verification rule.

[0100] In a possible design, the target verification rule can be converted into a regular expression, for example, "the insurance terms and premium rate text has been submitted" is converted into "`Product.term_element_table_path`match`\S`". Thus, the target verification rule can be described using a regular expression, a normative statement, which increases the convenience of rule verification.

[0101] Step 108, when the product risk association information of the detected subject matches the target verification rule, determine that the detected subject is a compliant subject; otherwise, determine that the detected subject is a non-compliant subject.

[0102] At this point, if the two match, it means that the subject being tested meets the requirements of the target verification rules and is a compliant subject. Conversely, if the two do not match, it means that the subject being tested does not meet the requirements of the target verification rules and is a non-compliant subject. As shown in the example in Table 1 below, for the tested subjects xxxx Property Insurance Co., Ltd. Beijing Middle-aged and Elderly Hospitalization Medical Insurance (Internet Exclusive) and xxxx Property Insurance Co., Ltd. Beijing Middle-aged and Elderly Hospitalization Medical Insurance, multiple product risk-related information and their corresponding verification results are distributed as follows. In the end, xxxx Property Insurance Co., Ltd. Beijing Middle-aged and Elderly Hospitalization Medical Insurance (Internet Exclusive) was judged as a compliant subject, and xxxx Property Insurance Co., Ltd. Beijing Middle-aged and Elderly Hospitalization Medical Insurance was judged as a non-compliant subject.

[0103] Table 1

[0104]

[0105] The above technical solution can automatically screen the applicable target verification rules based on the basic information of its products for businesses that provide property insurance certificates to users, and then automatically check whether its product risk-related information matches the target verification rules, thereby realizing the automated judgment of whether the business is compliant. Through this technical solution, the labor cost and time cost of compliance inspection of property insurance business are reduced, and the convenience and accuracy of compliance inspection of property insurance business are improved. While reducing the management cost of the business management party, it can also effectively enhance the effectiveness of compliance review of property insurance business, indirectly improve the experience of insured users, and more safely and effectively protect their property rights and interests.

[0106] In one embodiment of the present application, the preset verification rule database includes completeness rules, compliance rules and early warning rules, then before step 102, it also includes: obtaining the verification requirements of the detected subject, wherein the verification requirements are used to reflect the type of verification that the detected subject needs to perform, and the verification type includes one or more of completeness verification, compliance verification and early warning verification; step 104 includes: selecting a target verification rule in the preset verification rule database that is associated with the detected subject and compatible with the verification requirements.

[0107] In other words, the preset verification rule database can be divided into different types according to the verification requirements of the business. When faced with different verification requirements, the target verification rules can be screened only under the rule types that meet the verification requirements. This reduces the base number of rule screening, makes rule screening easier, and simplifies the overall process of compliance review.

[0108] In addition, if the target verification rule includes that the comprehensive product risk score of the inspected subject is less than or equal to a predetermined threshold, step 106 includes: determining a first risk parameter corresponding to each item of product risk association information of the inspected subject; based on the first risk parameter, determining a second risk parameter corresponding to each item of product risk association information of the inspected subject; constructing a risk degree matrix based on a risk parameter sequence obtained by arranging the first risk parameter and the second risk parameter from large to small, wherein the risk degree matrix is ​​used to reflect the risk level distribution of the inspected subject under the individual influence and interactive influence of each product risk association information; determining the comprehensive product risk score of the inspected subject based on the risk degree matrix and the risk parameter sequence; if the comprehensive product risk score is less than or equal to the predetermined threshold, determining that the product risk association information of the inspected subject matches the target verification rule, otherwise, determining that the product risk association information of the inspected subject does not match the target verification rule.

[0109] Specifically, each product risk association information of the detected subject is encoded to obtain a first risk parameter. Optionally, the first risk parameter can be normalized so that the values ​​of each dimension are at the same level, which is convenient for subsequent calculation and processing.

[0110] Next, the second risk parameter corresponding to each of the multiple product risk association information of the detected subject can be determined. Optionally, the mean of the first risk parameters of the multiple product risk association information can be set as the second risk parameter corresponding to the multiple product risk association information. Optionally, the square root of the product of the first risk parameters of the multiple product risk association information can be set as the second risk parameter corresponding to the multiple product risk association information. The second risk parameter reflects the comprehensive impact of the multiple product risk association information on the detected subject.

[0111] At this point, a risk matrix is ​​constructed based on a risk parameter sequence obtained by arranging the first risk parameter and the second risk parameter from large to small.

[0112] Specifically, the element of the first designated row and the first designated column in the risk degree matrix is ​​set to be the product of the risk parameter of the first designated order and the risk parameter of the second designated order in the risk parameter sequence. Then the element reflects the risk level of the detected subject under the combined influence of the product risk association information corresponding to the risk parameter of the first designated order and the multiple product risk association information corresponding to the risk parameter of the second designated order. The risk degree matrix reflects the distribution of the risk level of the detected subject under the individual influence and interactive influence of each product risk association information.

[0113] Finally, based on the risk matrix and the risk parameter sequence, the comprehensive risk score of the product of the detected subject is determined. If the comprehensive risk score of the product is less than or equal to the predetermined threshold, it is determined that the product risk association information of the detected subject matches the target verification rule; otherwise, it is determined that the product risk association information of the detected subject does not match the target verification rule.

[0114] The specific method for determining the comprehensive risk score of the product of the subject under test is as follows: obtaining the rank of the risk matrix as the third risk parameter; obtaining the variance of the risk parameter sequence as the fourth risk parameter; normalizing the third risk parameter and the fourth risk parameter, and averaging the results of the normalization, and using the average as the comprehensive risk score of the product of the subject under test. It can be seen that the comprehensive risk score of the product reflects the risk assignment of the subject under test under the premise of comprehensively considering the risk levels of the risk matrix and the risk parameter sequence.

[0115] The above technical solution can analyze the risk level of the inspected subject in a multi-dimensional and multi-directional manner based on multiple product risk-related information of the inspected subject, thereby improving the accuracy of risk detection of the inspected subject, providing further guarantees for the compliance review of property insurance, and helping to protect the property safety of property insurance customers.

[0116] The present application provides a data verification device, which includes:

[0117] An information acquisition unit, used to acquire basic product information and product risk-related information of a detected subject, wherein the detected subject is a business that provides property protection certificates to users, the basic product information includes the product category, sales channel, development method, operating area and insurance liability type of the business, and the product risk-related information is information that has an impact on the property protection capability of the business;

[0118] A verification rule determination unit, configured to determine a target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database;

[0119] A verification rule matching unit, used to determine whether the product risk association information of the detected subject matches the target verification rule;

[0120] The compliance subject determination unit is used to determine that the detected subject is a compliant subject when the product risk association information of the detected subject matches the target verification rule, otherwise, determine that the detected subject is a non-compliant subject.

[0121] In one embodiment of the present application, optionally, the device further includes:

[0122] A verification rule database setting unit is used to determine the corresponding verification rule of the single initial rule information based on the single initial rule information and the verification rule identification model in the initial rule information sample set before obtaining the basic product information and product risk association information of the inspected subject, and add the corresponding verification rule to the verification rule database, wherein the verification rule identification model is used to reflect the association relationship between the single initial rule information and the corresponding verification rule in the historical valid rule verification task.

[0123] In one embodiment of the present application, optionally, the verification rule determination unit includes:

[0124] The first execution unit is used to obtain an entity set of the preset verification rule database; based on the basic product information of the detected subject, determine a target entity associated with the detected subject in the entity set; and filter the verification rules with the target entity in the preset verification rule database as the target verification rules associated with the detected subject.

[0125] In one embodiment of the present application, optionally, the verification rule determination unit includes:

[0126] The second execution unit is used to determine the search keywords based on the basic product information of the detected subject; determine the effective rules hit by the search keywords in the preset verification rule database; determine the entities and entity relationships in the effective rules based on the effective rules and a preset large language model, wherein the preset large language model is used to reflect the association relationship between the description text of the effective rules and the entities and entity relationships in the effective rules; generate the target verification rules based on the entities and entity relationships in the effective rules.

[0127] In one embodiment of the present application, optionally, the preset verification rule database includes completeness rules, compliance rules and early warning rules, and the device further includes:

[0128] A verification requirement acquisition unit, used to acquire the verification requirement of the detected subject, wherein the verification requirement is used to reflect the type of verification that the detected subject needs to undergo, and the verification type includes one or more of completeness verification, compliance verification, and early warning verification;

[0129] The verification rule determination unit is used to select a target verification rule from the preset verification rule database that is associated with the detected subject and is compatible with the verification requirement.

[0130] In one embodiment of the present application, optionally, the verification rule matching unit is used to perform at least one of the following detection methods:

[0131] Determine whether the numerical value in the product risk association information matches the numerical range restriction information in the target verification rule; determine whether the segmented words in the product risk association information hit all the keywords in the target verification rule; detect whether the product risk association information contains specified signature information through OCR recognition; detect whether the product risk association information includes a specified file.

[0132] In one embodiment of the present application, optionally, the target verification rule includes that the comprehensive risk score of the product of the detected subject is less than or equal to a predetermined threshold, and the verification rule matching unit includes:

[0133] A first risk parameter determination unit, used to determine a first risk parameter corresponding to each item of product risk association information of the detected subject;

[0134] A second risk parameter determination unit, configured to determine, based on the first risk parameter, a second risk parameter corresponding to each of the plurality of product risk association information of the detected subject;

[0135] a matrix construction unit, configured to construct a risk degree matrix based on a risk parameter sequence obtained by arranging the first risk parameter and the second risk parameter from large to small, wherein the risk degree matrix is ​​used to reflect the risk level distribution of the detected subject under the individual influence and interactive influence of each product risk association information;

[0136] A risk score calculation unit, used to determine the comprehensive risk score of the product of the detected subject based on the risk matrix and the risk parameter sequence;

[0137] The third execution unit is used to determine that the product risk association information of the detected subject matches the target verification rule if the comprehensive risk score of the product is less than or equal to the predetermined threshold; otherwise, determine that the product risk association information of the detected subject does not match the target verification rule.

[0138] In one embodiment of the present application, optionally, the matrix construction unit is used to: set the elements of the first specified row and the first specified column in the risk matrix to be the product of the risk parameters of the first specified order and the risk parameters of the second specified order in the risk parameter sequence.

[0139] In one embodiment of the present application, optionally, the risk score calculation unit is used to: obtain the rank of the risk matrix as a third risk parameter; obtain the variance of the risk parameter sequence as a fourth risk parameter; normalize the third risk parameter and the fourth risk parameter, and average the results of the normalization, and use the average as the comprehensive risk score of the product of the inspected subject.

[0140] The device uses any one of the solutions described in the above embodiments, and therefore has all the above technical effects, which will not be repeated here.

[0141] In addition, in one embodiment, the present application provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 2As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client via a network connection. When the computer program is executed by the processor, the method described in any of the above embodiments can be implemented.

[0142] In one embodiment, the present application further provides a computer device, which may be a client, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When the computer program is executed by the processor, the method described in any of the above embodiments can be implemented.

[0143] Any of the above-mentioned computer devices in the embodiments of the present application may exist in various forms, including but not limited to:

[0144] (1) Mobile communication devices: These devices are characterized by having mobile communication functions and their main purpose is to provide voice and data communications. These terminals include: smart phones (such as iPhone), multimedia phones, functional phones, and low-end phones.

[0145] (2) Ultra-mobile personal computer devices: These devices fall into the category of personal computers, have computing and processing capabilities, and generally also have mobile Internet access features. These terminals include: PDA, MID and UMPC devices, such as iPad.

[0146] (3) Portable entertainment devices: These devices can display and play multimedia content. They include audio and video players (such as iPods), handheld game consoles, e-books, as well as smart toys, wearable devices, and portable car navigation devices.

[0147] (4) Server: A device that provides computing services. The server consists of a processor, hard disk, memory, system bus, etc. The server is similar to a general computer architecture, but because it needs to provide highly reliable services, it has higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.

[0148] (5) Other electronic devices with data interaction functions.

[0149] In addition, an embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to perform the following steps:

[0150] Obtaining basic product information and product risk-related information of the subject being tested, wherein the subject being tested is a business that provides property protection certificates to users, the basic product information includes the product category, sales channel, development method, operating area and insurance liability type of the business, and the product risk-related information is information that has an impact on the property protection capability of the business;

[0151] Based on the basic product information of the detected subject and a preset verification rule database, determining a target verification rule associated with the detected subject;

[0152] Determining whether the product risk association information of the detected subject matches the target verification rule;

[0153] In the case where the product risk association information of the detected subject matches the target verification rule, the detected subject is determined to be a compliant subject; otherwise, the detected subject is determined to be a non-compliant subject.

[0154] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant description in the aforementioned method embodiment. In order to avoid repetition, they will not be described one by one here.

[0155] The above describes in detail the technical solution of the present application in conjunction with the accompanying drawings. Through the technical solution of the present application, for businesses that provide property protection certificates to users, the applicable target verification rules can be automatically screened based on their basic product information, and then their product risk-related information can be automatically checked to see if it matches the target verification rules, thereby achieving automated judgment on whether the business is compliant. Through this technical solution, the labor cost and time cost of compliance inspections for property insurance business are reduced, and the convenience and accuracy of compliance inspections for property insurance business are improved. While reducing the management costs of business managers, it can also effectively enhance the effectiveness of compliance reviews of property insurance business, indirectly improving the experience of policyholders and more safely and effectively protecting their property rights and interests.

[0156] It should be understood that although the terms first, second, etc. may be used to describe risk parameters in the embodiments of the present application, these risk parameters should not be limited to these terms. These terms are only used to distinguish risk parameters from each other. For example, without departing from the scope of the embodiments of the present application, the first risk parameter may also be referred to as the second risk parameter, and similarly, the second risk parameter may also be referred to as the first risk parameter.

[0157] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0158] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0159] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0160] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0161] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0162] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A data verification method, characterized in that: include: Obtaining basic product information and product risk-related information of the subject being tested, wherein the subject being tested is a business that provides property protection certificates to users, the basic product information includes the product category, sales channel, development method, operating area and insurance liability type of the business, and the product risk-related information is information that has an impact on the property protection capability of the business; Based on the basic product information of the detected subject and a preset verification rule database, determining a target verification rule associated with the detected subject; Determining whether the product risk association information of the detected subject matches the target verification rule; In the case where the product risk association information of the detected subject matches the target verification rule, the detected subject is determined to be a compliant subject; otherwise, the detected subject is determined to be a non-compliant subject.

2. The data verification method according to claim 1, characterized in that: Before obtaining the basic product information and product risk association information of the subject to be tested, the method further includes: Based on the single initial rule information and the verification rule identification model in the initial rule information sample set, the corresponding verification rule of the single initial rule information is determined, and the corresponding verification rule is added to the verification rule database, wherein the verification rule identification model is used to reflect the association between the single initial rule information and the corresponding verification rule in the historical valid rule verification task.

3. The data verification method according to claim 1, characterized in that: The determining of the target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database includes: Obtaining an entity set of the preset verification rule database; Based on the basic product information of the detected subject, determining a target entity associated with the detected subject in the entity set; The verification rule having the target entity is screened in the preset verification rule database as the target verification rule associated with the detected subject.

4. The data verification method according to claim 1, characterized in that: The determining of the target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database includes: Determine search keywords based on the basic product information of the detected subject; Determining the effective rules for the search keyword hits in the preset verification rule database; Based on the effective rules and a preset large language model, determining entities and inter-entity relationships in the effective rules, wherein the preset large language model is used to reflect the association relationship between the description text of the effective rules and the entities and inter-entity relationships in the effective rules; The target verification rules are generated based on the entities and the relationships between entities in the effective rules.

5. The data verification method according to any one of claims 1 to 4, characterized in that: The preset verification rule database includes completeness rules, compliance rules and early warning rules. Before obtaining the basic product information and product risk association information of the subject to be tested, the method further includes: Acquire the verification requirements of the detected subject, wherein the verification requirements are used to reflect the type of verification that the detected subject needs to undergo, and the verification type includes one or more of completeness verification, compliance verification, and early warning verification; The determining of the target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database includes: A target verification rule that is associated with the detected subject and matches the verification requirement is selected from the preset verification rule database.

6. The data verification method according to claim 1, characterized in that: The determining whether the product risk association information of the detected subject matches the target verification rule includes at least one of the following detection methods: Determining whether the value in the product risk association information matches the value range restriction information in the target verification rule; Determine whether the segmented words in the product risk association information hit all the keywords in the target verification rules; Detecting whether the product risk association information contains designated signature information by means of OCR recognition; Check whether the product risk association information includes a specified file.

7. The data verification method according to claim 1, characterized in that: The target verification rule includes that if the comprehensive risk score of the product of the detected subject is less than or equal to a predetermined threshold, then The determining whether the product risk association information of the detected subject matches the target verification rule includes: Determine a first risk parameter corresponding to each item of product risk association information of the detected subject; Based on the first risk parameter, determine a second risk parameter corresponding to each plurality of product risk association information of the detected subject; Based on the risk parameter sequence obtained by arranging the first risk parameter and the second risk parameter from large to small, a risk degree matrix is ​​constructed, wherein the risk degree matrix is ​​used to reflect the risk level distribution of the detected subject under the individual influence and interactive influence of each product risk association information; Determining a comprehensive risk score of the product of the inspected subject based on the risk matrix and the risk parameter sequence; If the comprehensive risk score of the product is less than or equal to the predetermined threshold, it is determined that the product risk association information of the detected subject matches the target verification rule; otherwise, it is determined that the product risk association information of the detected subject does not match the target verification rule.

8. The data verification method according to claim 7, characterized in that: The step of constructing a risk matrix based on a risk parameter sequence obtained by arranging the first risk parameter and the second risk parameter from large to small includes: The element of the first designated row and the first designated column in the risk matrix is ​​set to be the product of the risk parameter of the first designated order and the risk parameter of the second designated order in the risk parameter sequence.

9. The data verification method according to claim 7, characterized in that: The step of determining the comprehensive risk score of the product of the detected subject based on the risk matrix and the risk parameter sequence includes: Obtaining the rank of the risk matrix as a third risk parameter; Obtaining the variance of the risk parameter sequence as a fourth risk parameter; The third risk parameter and the fourth risk parameter are normalized, and the results of the normalization are averaged, and the average is used as the comprehensive risk score of the product of the detected subject.

10. A data verification device, characterized in that: include: An information acquisition unit, used to acquire basic product information and product risk-related information of a detected subject, wherein the detected subject is a business that provides property protection certificates to users, the basic product information includes the product category, sales channel, development method, operating area and insurance liability type of the business, and the product risk-related information is information that has an impact on the property protection capability of the business; A verification rule determination unit, configured to determine a target verification rule associated with the detected subject based on the basic product information of the detected subject and a preset verification rule database; A verification rule matching unit, used to determine whether the product risk association information of the detected subject matches the target verification rule; The compliance subject determination unit is used to determine that the detected subject is a compliant subject when the product risk association information of the detected subject matches the target verification rule, otherwise, determine that the detected subject is a non-compliant subject.

11. A computer device, characterized in that: include: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions executable by the at least one processor, wherein the instructions are configured to execute the method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: Computer executable instructions are stored, and the computer executable instructions are used to execute the method according to any one of claims 1 to 9.

Images