CN114581251A

CN114581251A - Data verification method and device, computer equipment and computer readable storage medium

Info

Publication number: CN114581251A
Application number: CN202210207785.4A
Authority: CN
Inventors: 骆世越; 徐定伟
Original assignee: Shenzhen One Ledger Science And Technology Service Co ltd
Current assignee: Shenzhen One Ledger Science And Technology Service Co ltd
Priority date: 2022-03-03
Filing date: 2022-03-03
Publication date: 2022-06-03

Abstract

The application discloses a data verification method, a data verification device, computer equipment and a computer readable storage medium, relates to the technical field of internet, and can realize complete analysis of associated data so as to improve the accuracy of a target case verification result. The method comprises the following steps: acquiring a plurality of sample cases from a historical verification database, and determining a verification result corresponding to each sample case according to a plurality of sample associated data corresponding to each sample case; comparing the verification result with the verification result corresponding to the sample case, and counting the number of correct samples with the verification result consistent with the content of the verification result; when the ratio of the number of correct samples to the total number of samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold, obtaining a verification rule model; and acquiring a target case waiting for data verification in the case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case.

Description

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

Technical Field

The present application relates to the field of internet technologies, and in particular, to a data verification method and apparatus, a computer device, and a computer-readable storage medium.

Background

With the continuous progress of the internet technology and the gradual development of the insurance industry, the insurance policy amount begins to rise, and in order to strictly control the insurance policy risk, the insurance platform adopts a manual auditing mode to carry out the underwriting on the insurance policy generated by the insurance platform.

In the related technology, after receiving a case needing to be underwritten, a worker of manual underwriting inquires related information such as self-underwriting data, a disease label and the like of the case according to client information corresponding to the case, and gives an underwriting conclusion suggestion according to the related information.

In implementing the present invention, the applicant has found that the related art has at least the following problems:

with the increase of business volume, cases needing manual review are more and more, the processing period is longer and longer, and the specialities of reviewers are also good and irregular, so that the analysis of the correlation factors of the reported cases by manual review is less, the issuing timeliness of the letters is low, and the accuracy of the verification result is low.

Disclosure of Invention

In view of the above, the present application provides a data verification method, an apparatus, a computer device, and a computer readable storage medium, and mainly aims to solve the problems that the current cases requiring manual review are more and more, the processing period is longer and longer, and the specialities of reviewers are also good and varied, which results in less analysis of the association factors of the reported cases by manual review, low timeliness of letter issuing, and low accuracy of verification results.

According to a first aspect of the present application, there is provided a data verification method, including:

obtaining a plurality of sample cases from a historical verification database, and determining a verification result corresponding to each sample case according to a plurality of sample associated data corresponding to each sample case, wherein the sample associated data is business data generated by a sample client corresponding to the sample case on a platform;

comparing the verification result with the auditing result corresponding to the sample case, and counting the number of correct samples with the verification result consistent with the content of the auditing result;

when the ratio of the number of the correct samples to the total number of the samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold, obtaining a check rule model;

and acquiring a target case waiting for data verification in a case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case.

Optionally, the determining, according to a plurality of sample associated data corresponding to each sample case in the plurality of sample cases, a verification result of each sample case includes:

reading text information of the sample case, and extracting sample client information and sample audit data corresponding to the sample case from the text information, wherein the sample client information is used for indicating a sample client corresponding to the sample case, and the sample audit data is used for indicating a case type corresponding to the sample case;

calling an external system connection interface, identifying client tags of all service data in an external service database, determining a plurality of specified client tags with tag contents consistent with the sample client information, and taking a plurality of service data corresponding to the specified client tags as the sample associated data;

inquiring a plurality of check rule items corresponding to the sample audit data in a check rule database, and performing data check on the plurality of sample associated data according to the plurality of check rule items to obtain a plurality of sub-check results;

in the plurality of sub-verification results, dividing at least one sub-verification result with consistent result content into the same result group to obtain a plurality of result groups;

determining a total result ratio corresponding to each result group in the plurality of result groups to obtain a plurality of total result ratios;

sequencing the total result ratios, determining a designated total result ratio with the ranking meeting a preset condition, and taking a sub-verification result corresponding to the designated total result ratio as a verification result corresponding to the sample case;

for each sample case in the plurality of sample cases, obtaining a plurality of sample associated data corresponding to each sample case, determining a verification result of each sample case, and obtaining a plurality of verification results.

Optionally, the performing data verification on the multiple sample associated data according to the multiple verification rule items to obtain multiple sub-verification results includes:

for each sample associated data in the plurality of sample associated data, determining a data category of the sample associated data, and inquiring a first designated check rule item with an item identification consistent with the data category in the plurality of check rule items;

verifying the sample associated data according to a verification index corresponding to the first specified verification rule item, and extracting a verification result hit by the sample associated data as a verification result corresponding to the sample associated data;

and verifying each sample associated data in the plurality of sample associated data to obtain a plurality of verification results.

Optionally, the determining a total result ratio corresponding to each result group in the plurality of result groups to obtain a plurality of total result ratios includes:

for each result set of the plurality of result sets, identifying at least one verification result contained in the result set;

determining a second specified verification rule item corresponding to each sub-verification result in the at least one sub-verification result, and taking the proportion weight corresponding to the second specified verification rule item as the result proportion corresponding to each sub-verification result to obtain at least one result proportion corresponding to the at least one sub-verification result;

adding the at least one result ratio to obtain a total result ratio corresponding to each result group;

and calculating the total result ratio corresponding to each result group in the plurality of result groups to obtain the plurality of total result ratios.

Optionally, the comparing the multiple verification results with the audit results corresponding to the multiple sample cases, and counting the number of correct samples with the verification results consistent with the audit results in the multiple verification results includes:

identifying an audit result corresponding to each sample case in the plurality of sample cases, and comparing the audit result corresponding to each sample case with a verification result corresponding to each sample case;

determining a target sample case with a verification result consistent with an auditing result in the plurality of sample cases, and adding an auditing label for indicating correct content to the target sample case;

counting the number of target sample cases added with audit tags for indicating the correct contents to obtain the correct sample number.

Optionally, after the target case waiting for data verification is acquired from the case database, and the target case is input to the verification rule model for data verification, and a target verification result of the target case is obtained, the method further includes:

determining a verification completion time point of the target verification result, and continuously counting the time interval between the current time point and the verification completion time point;

and when the time interval reaches a preset time interval, acquiring a new target case from the case database again, inputting the new target case into the verification rule model for data verification, and obtaining a target verification result of the new target case.

Optionally, the method further comprises:

sending the verification result to an audit terminal for displaying, and acquiring an uploaded audit result of the audit terminal;

when the verification result indicates that the verification result is changed, extracting the verification result and the modification reason uploaded by the verification terminal, marking the target case information by adopting the verification result, and storing the marked target case information to the historical verification database;

extracting a target verification rule item and a modification index from the modification reason, and inquiring the target verification index corresponding to the target verification rule item in a verification rule database;

and updating the target verification index by adopting the modification index, and storing the updated target verification index into the verification rule database.

According to a second aspect of the present application, there is provided a data verification apparatus, comprising:

the system comprises a determining module, a verification module and a verification module, wherein the determining module is used for acquiring a plurality of sample cases from a historical verification database, and determining a verification result corresponding to each sample case according to a plurality of sample associated data corresponding to each sample case, and the sample associated data is business data generated by a sample client corresponding to the sample case on a platform;

the comparison module is used for comparing the verification result with the auditing result corresponding to the sample case and counting the number of correct samples with the verification result consistent with the content of the auditing result;

the calculation module is used for obtaining a check rule model when the ratio of the number of the correct samples to the total number of the samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold;

and the verification module is used for acquiring a target case waiting for data verification in a case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case.

Optionally, the determining module is configured to read text type information of the sample case, and extract sample client information and sample audit data corresponding to the sample case from the text type information, where the sample client information is used to indicate a sample client corresponding to the sample case, and the sample audit data is used to indicate a case type corresponding to the sample case; calling an external system connection interface, identifying client tags of all service data in an external service database, determining a plurality of specified client tags with tag contents consistent with the sample client information, and taking a plurality of service data corresponding to the specified client tags as the sample associated data; inquiring a plurality of check rule items corresponding to the sample audit data in a check rule database, and performing data check on the plurality of sample associated data according to the plurality of check rule items to obtain a plurality of sub-check results; in the plurality of sub-verification results, dividing at least one sub-verification result with consistent result content into the same result group to obtain a plurality of result groups; determining a total result ratio corresponding to each result group in the plurality of result groups to obtain a plurality of total result ratios; sequencing the total result ratios, determining a designated total result ratio with the ranking meeting a preset condition, and taking a sub-verification result corresponding to the designated total result ratio as a verification result corresponding to the sample case; for each sample case in the plurality of sample cases, obtaining a plurality of sample associated data corresponding to each sample case, determining a verification result of each sample case, and obtaining a plurality of verification results.

Optionally, the determining module is configured to determine, for each sample associated data in the plurality of sample associated data, a data category of the sample associated data, and query an item in the plurality of check rule items to identify a first specified check rule item consistent with the data category; verifying the sample associated data according to a verification index corresponding to the first specified verification rule item, and extracting a verification result hit by the sample associated data as a sub-verification result corresponding to the sample associated data; and verifying each sample associated data in the plurality of sample associated data to obtain a plurality of sub-verification results.

Optionally, the determining module is configured to identify, for each result group of the plurality of result groups, at least one sub-verification result included in the result group; determining a second specified verification rule item corresponding to each sub-verification result in the at least one sub-verification result, and taking the proportion weight corresponding to the second specified verification rule item as the result proportion corresponding to each sub-verification result to obtain at least one result proportion corresponding to the at least one sub-verification result; adding the at least one result ratio to obtain a total result ratio corresponding to each result group; and calculating the total result ratio corresponding to each result group in the plurality of result groups to obtain the plurality of total result ratios.

Optionally, the comparison module is configured to identify an audit result corresponding to each sample case in the plurality of sample cases, and compare the audit result corresponding to each sample case with the verification result corresponding to each sample case; determining a target sample case with a verification result consistent with an auditing result in the plurality of sample cases, and adding an auditing label for indicating correct content to the target sample case; counting the number of target sample cases added with audit tags for indicating the correct contents to obtain the correct sample number.

Optionally, the apparatus further comprises:

the statistical module is used for determining a verification completion time point of the target verification result and continuously counting the time interval between the current time point and the verification completion time point;

and the verification module is used for acquiring a new target case from the case database again when the time interval reaches a preset time interval, inputting the new target case into the verification rule model for data verification, and obtaining a target verification result of the new target case.

Optionally, the apparatus further comprises:

the display module is used for sending the verification result to an audit terminal for displaying and acquiring the uploaded audit result of the audit terminal;

the marking module is used for extracting the verification result and the modification reason uploaded by the verification terminal when the verification result indicates to change the verification result, marking the target case information by adopting the verification result, and storing the marked target case information to the historical verification database;

the extraction module is used for extracting a target verification rule item and a modification index from the modification reason and inquiring the target verification index corresponding to the target verification rule item in a verification rule database;

and the storage module is used for updating the target verification index by adopting the modification index and storing the updated target verification index into the verification rule database.

According to a third aspect of the present application, there is provided a computer device comprising a memory storing a computer program and a processor implementing the steps of the method of any of the first aspects when the computer program is executed.

According to a fourth aspect of the present application, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of any of the first aspects described above.

By means of the technical scheme, the data verification method, the data verification device, the computer equipment and the computer readable storage medium are provided. And then, comparing the verification result with the auditing result corresponding to the sample case, counting the number of correct samples with the verification result consistent with the content of the auditing result, and obtaining a verification rule model when the ratio of the number of the correct samples to the total number of the samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold value. And finally, acquiring a target case waiting for data verification in the case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case. The related data stored in the platform database of the target case is exhausted by utilizing the computing power of the computer, and a verification rule model is generated by comparing the related data with the corresponding indexes, so that the complete analysis of the related data is realized, and the accuracy of the verification result of the target case is further improved.

The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.

Drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the application. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

fig. 1 shows a schematic flow chart of a data verification method provided in an embodiment of the present application;

fig. 2 is a schematic flow chart illustrating a data verification method provided in an embodiment of the present application;

fig. 3 shows a schematic structural diagram of a data verification apparatus provided in an embodiment of the present application;

fig. 4 shows a schematic device structure diagram of a computer apparatus according to an embodiment of the present application.

Detailed Description

Exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

An embodiment of the present application provides a data verification method, as shown in fig. 1, the method includes:

101. the method comprises the steps of obtaining a plurality of sample cases from a historical verification database, and determining a verification result corresponding to each sample case according to a plurality of sample associated data corresponding to each sample case, wherein the sample associated data is business data generated by a sample client corresponding to the sample case on a platform.

102. And comparing the verification result with the auditing result corresponding to the sample case, and counting the number of correct samples with the verification result consistent with the content of the auditing result.

103. And when the ratio of the number of the correct samples to the total number of the samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold, obtaining a verification rule model.

104. And acquiring a target case waiting for data verification in the case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case.

According to the method provided by the embodiment of the application, a plurality of sample cases are obtained from a historical verification database, and the verification result corresponding to each sample case is determined according to a plurality of sample associated data corresponding to each sample case. And then, comparing the checking result with the auditing result corresponding to the sample case, counting the number of correct samples with the same content of the checking result and the auditing result, and obtaining a checking rule model when the ratio of the number of the correct samples to the total number of the samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold value. And finally, acquiring a target case waiting for data verification in the case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case. The method has the advantages that the computing power of a computer is utilized, the associated data stored in the platform database by the target case are exhausted, the associated data are compared with the corresponding indexes to generate a verification rule model, the complete analysis of the associated data is realized, and the accuracy of the verification result of the target case is improved.

An embodiment of the present application provides a data verification method, as shown in fig. 2, the method includes:

201. obtaining a plurality of sample cases from a historical verification database, and determining a verification result corresponding to each sample case according to a plurality of sample associated data corresponding to each sample case.

With the continuous progress of the internet technology and the gradual development of the insurance industry, the insurance policy amount begins to rise, and in order to strictly control the insurance policy risk, the insurance platform adopts a manual auditing mode to carry out the underwriting on the insurance policy generated by the insurance platform. At present, after receiving a case needing to be underwritten, a worker for manual underwriting inquires relevant information such as self-underwriting data, a disease label and the like of the case according to client information corresponding to the case, and gives an underwriting conclusion suggestion according to the relevant information. However, the applicant realizes that with the increase of the business volume, cases needing manual auditing are more and more, the processing period is longer and longer, and the specialities of auditors are also good and uneven, so that the analysis of the association factors of reported cases by manual auditing is less, the issuing timeliness of the cases is low, and the accuracy of the verification result is low.

Therefore, according to the data verification method, the data verification device, the computer equipment and the computer-readable storage medium, a plurality of sample cases are firstly obtained from a historical verification database, and the verification result corresponding to each sample case is determined according to a plurality of sample associated data corresponding to each sample case. And then, comparing the checking result with the auditing result corresponding to the sample case, counting the number of correct samples with the same content of the checking result and the auditing result, and obtaining a checking rule model when the ratio of the number of the correct samples to the total number of the samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold value. And finally, acquiring a target case waiting for data verification in the case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case. The related data stored in the platform database of the target case is exhausted by utilizing the computing power of the computer, and a verification rule model is generated by comparing the related data with the corresponding indexes, so that the complete analysis of the related data is realized, and the accuracy of the verification result of the target case is further improved.

The method and the device are applied to a human nuclear system, a verification rule model is established based on the sample case with the known auditing result, and the target verification result is generated by inputting the target sample case to be verified into the verification rule model. In addition, the human-nuclear system can finally send the target verification result to the verification terminal for displaying, and continuously trains the verification rule model according to the verification result of the related staff on the target verification result, so that the verification rule model with higher accuracy is obtained.

In the embodiment of the application, the fact that the platform staff can input the auditing result obtained by manual auditing into the historical verifying database is considered, and therefore a large amount of historical case information which is already manually audited is stored in the historical verifying database. Specifically, the system firstly obtains a plurality of sample cases from a historical verification database, and determines a plurality of verification results corresponding to the plurality of sample cases according to a plurality of sample associated data corresponding to each sample case in the plurality of sample cases. For each sample case in a plurality of sample cases, the specific process of obtaining the verification result is divided into the following steps:

step one, identifying each sample case, and determining a plurality of sample associated data corresponding to the sample case.

In this step, for each sample case in the plurality of sample cases, the text type information of the sample case may be read, and sample customer information indicating a sample customer corresponding to the sample case and sample audit data indicating a case type corresponding to the sample case are extracted from the text type information. For example, a customer number, a customer name, a customer sex, a customer age, an identification number, and the like are extracted from the text information as sample customer information, and contents such as a policy type and a case initiation cause are extracted as sample audit data. In the actual application process, the content types extracted from the sample client information and the sample audit data can adopt the default content types of the system, and can also be set by related workers according to the actual operation conditions. And then calling an external system connection interface, and identifying the client tags of all the service data in an external service database. In all the client tags, a plurality of designated client tags whose tag contents match the sample client information are specified, and a plurality of pieces of business data corresponding to the plurality of designated client tags are set as a plurality of pieces of sample-related data. It should be noted that the sample-related data is used to indicate all business behavior data generated by the sample client on the insurance platform, for example, self-checking data, wind control tags, security records, past disease records, claim settlement records, and the like, and the type of the sample-related data is not specifically limited in the present application. In addition, because different types of associated data are stored in the underwriting databases corresponding to different systems of the platform, after the system acquires the sample client information, an external system link interface needs to be called to connect with other systems of the platform. And further inquiring sample associated data corresponding to the sample customer information in an underwriting database corresponding to other systems according to the sample customer information. Specifically, the system queries the past artificial underwriting information of the client from the human-based system according to the sample client information, namely the insured person information, acquires the past wind-controlled data from the wind-controlled intelligent underwriting system, acquires the past data from the self-checking system, queries the past insurance-purchasing record of the client from the insurance-purchasing system, and acquires the past insurance-purchasing record of the client from the insurance-purchasing system and the insurance-paying system. In the actual application process, the database corresponding to each system is considered to add corresponding customer labels, such as customer number labels, basic information labels and the like, when storing the behavior data, so that the system can retrieve the corresponding customer labels to compare with the customer numbers stored in the sample customer information, and extract the behavior data corresponding to the sample customer information.

The method is connected with an external system, and utilizes the computing power of a computer to obtain all sample associated data of a sample client, so as to provide a complete data base for generating a verification result based on the subsequent analysis of the sample associated data.

And secondly, inquiring a verification rule corresponding to each sample case based on the sample audit data, and verifying the associated data of the plurality of samples by adopting the verification rule to obtain a sub-verification result corresponding to the associated data of each sample.

In this step, the system firstly queries a plurality of check rule items corresponding to the sample audit data in the check rule database, and performs data check on the plurality of sample associated data according to the plurality of check rule items to obtain a plurality of sub-check results. It should be noted that the platform may upload the underwriting rule indicated by the text information to the system, and the system identifies the underwriting rule, determines the case type indicated in the underwriting rule, and adds a label to the underwriting rule using the case type.

In the practical application process, for each sample associated data in the plurality of sample associated data, the data category of the sample associated data is determined, such as self-checking data, a wind control label, a preservation record, a past disease record, a claim settlement record and the like. The query item identifies a first specified check rule item consistent with the data category in the plurality of check rule items. And then, verifying the sample associated data according to the verification index corresponding to the first specified verification rule item, extracting a verification result hit by the sample associated data as a sub-verification result corresponding to the sample associated data, if so, judging whether the self-core passes, and if not, indicating that the sub-verification result corresponding to the sample associated data is 'charging'. And finally, verifying each sample associated data in the plurality of sample associated data to obtain a plurality of sub-verification results.

And step three, determining a verification result based on the plurality of sub-verification results.

In fact, among the plurality of sub-verification results, at least one sub-verification result with consistent result content is divided into the same result group to obtain a plurality of result groups, for example, the sub-verification result with the conclusion of "charging" is divided into the same result group, the sub-verification result with the conclusion of "excluding" is divided into the same result group, and the like. Then, the total result ratio corresponding to each result group in the plurality of result groups is determined, and a plurality of total result ratios are obtained. Specifically, for each result group in the plurality of result groups, at least one verification result included in the result group is identified, and a second specified verification rule item corresponding to each verification result in the at least one verification result is determined. And taking the proportion weight corresponding to the second specified verification rule item as the result proportion corresponding to each verification result to obtain at least one result proportion corresponding to at least one verification result. And adding at least one result ratio to obtain a total result ratio corresponding to each result group, and calculating the total result ratio corresponding to each result group in the plurality of result groups to obtain a plurality of total result ratios. And then, sequencing the total result ratios, determining the designated total result ratio of which the rank meets the preset condition, and taking the sub-verification result corresponding to the designated total result ratio as the verification result corresponding to the sample case. For example, if the sample audit data sent to the system by the client a indicates that the policy type is the claim service type, the system acquires the underwriting rule corresponding to the claim service type, if the self-approval fails, outputs "charging", the percentage weight is 20%, the wind control label displays that the occurrence of an overlarge disease exceeds the claim threshold value, outputs "charging", the percentage 50%, the wind control label indicates that the information is not true, outputs "refusal to pay", the percentage 10%, the preservation record indicates that the information is not, outputs the normal claim percentage 10%, the claim record indicates that the information is not, outputs the normal claim percentage 5%, the past medical history record indicates that the information is not, outputs the normal claim percentage 5%, and finally outputs the audit result as "charging" according to the percentage ranking.

According to the sample client information stored in the sample case, the previous associated data of the sample client, including self-checking data, disease labels, past security, past claims and other associated information, is found through model training, and the checking result corresponding to the sample case is obtained, so that the analysis degree of the human checking system on the associated data is greatly enhanced, and the efficiency and the accuracy of generating the checking result are further improved.

202. And comparing the verification result with the verification result corresponding to the sample case, and counting the number of correct samples with the verification result consistent with the content of the verification result.

In the embodiment of the application, the verification result corresponding to each sample case is compared with the corresponding audit result, and whether the verification result generated by the verification rule model is consistent with the audit result or not is judged, that is, if the contents are consistent, the verification result is correct, and if the contents are inconsistent, the verification result is wrong.

Specifically, an audit result corresponding to each sample case in the plurality of sample cases is identified, and the audit result corresponding to each sample case is compared with a verification result corresponding to each sample case. And determining a target sample case with a verification result consistent with the auditing result in the plurality of sample cases, and adding an auditing label for indicating correct content to the target sample case. Further, the number of target sample cases added with the audit tags indicating that the content is correct is counted to obtain the correct sample number.

Through the steps, the audit tag indicating the correct content is added to the sample cases with the verification results consistent with the audit results, the number of the sample cases with the correct verification results is continuously counted through identifying the audit tag, and therefore the accuracy rate of the verification rule model for generating the verification results is calculated based on the number of the correct sample cases.

203. And when the ratio of the number of the correct samples to the total number of the samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold, obtaining a verification rule model.

In the embodiment of the application, the accuracy of the verification rule model for generating the verification result is obtained by calculating the ratio of the number of correct samples to the total number of samples of the plurality of sample cases, and when the accuracy is greater than or equal to the preset accuracy threshold, the verification rule model can output the correct verification result to the greatest extent. When the accuracy is smaller than the preset accuracy threshold, the parameters of the check rule model need to be adjusted, and the check rule model continues to be trained.

Specifically, the case type of the sample case with the wrong verification can be identified, a plurality of verification rule items corresponding to the case type are inquired, and the finally output verification result is adjusted by adjusting the result proportion parameters corresponding to different verification rule items. And extracting the information of the plurality of sample cases again, inputting the information into the verification rule model, outputting a verification result, determining the case type of the sample case with the wrong verification, and continuously adjusting the model parameters of the verification rule model until the accuracy of the generated verification result reaches a preset accuracy threshold.

Through the steps, a plurality of sample cases are continuously extracted for verification, model parameters of the verification rule model are updated, model training is carried out on the verification rule model, and finally training of the verification rule model is completed when the accuracy reaches a preset accuracy threshold value, so that the verification rule model is obtained.

204. And acquiring a target case waiting for data verification in the case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case.

In the embodiment of the present application, the system acquires the target case to be verified in the case database, inputs the target me's to the verification rule model for data verification, and outputs the target verification result of the target case, so that the relevant staff can perform case audit based on the target verification result.

The case database contains all the case information to be checked and maintained which is successfully reported, and data contents which are irrelevant to checking, such as home addresses, telephone numbers and the like, may exist in the target case information sent to the human-checking system. Therefore, the system needs to perform data cleaning on the acquired target case, and only the data content related to verification is reserved. Specifically, based on the natural language processing technology, target case information is identified, data indicating client identity information is extracted as target client information, and data indicating a case type, such as insured person information, a case type, policy information, and the like, is extracted as target audit data. Further, in practice, the system may set a time interval, determine a verification completion time point at which the target verification result is obtained, and continuously count the time interval between the current time point and the verification completion time point. And when the time interval reaches the preset time interval, acquiring a new target case from the case database again, inputting the new target case into the verification rule model for data verification, and obtaining a target verification result of the new target case. It should be noted that the system may set the target number of cases to be grabbed, grab the target number of cases each time, verify the cases, count time from the completion of the last target case, and when the time interval reaches the preset time interval, re-locate the target number of cases in the case database.

In another implementation scenario, after the verification rule model outputs the verification result, the system sends the generated verification result to the auditing terminal for display, so that a worker at the auditing terminal can perform manual auditing on the verification result. After the auditing is completed, the relevant staff need to upload the auditing result to the system, and then the system identifies the auditing result. And when the verification result indicates that the verification result is changed, extracting the verification result and the modification reason uploaded by the verification terminal, marking the target case information by adopting the verification result, and storing the marked target case information into a historical verification database. Further, extracting a target verification rule item and a modification index from the modification reason, inquiring the target verification index corresponding to the target verification rule item in a verification rule database, updating the target verification index by adopting the modification index, and storing the updated target verification index into the verification rule database. For example, if the probability of 80% for the human case that the auditor checks the claim has the security record is 80%, the 'refusal' of the automatic human checking conclusion is changed into 'charging', and the modification instruction indicates that the security record is not enough to refuse, the system automatically modifies the result proportion of the security record in the corresponding case of the claim in the scene so as to generate a more accurate verification result next time. In addition, when the verification result indicates that the verification result is confirmed, the verification result is adopted to mark the target case information, and the marked target case information is stored in the historical verification database.

In the practical application process, the system is provided with a manual auditing and reporting interface with a uniform caliber, and the case report meeting the requirement of sending a human core is sent to the system by calling the interface. The system stores the target case information into the case database, generates case reporting success reminding information and sends the case reporting success reminding information to an external core sending system to remind that the case reporting is successful.

According to the method provided by the embodiment of the application, a plurality of sample cases are obtained from a historical verification database, and the verification result corresponding to each sample case is determined according to a plurality of sample associated data corresponding to each sample case. And then, comparing the verification result with the auditing result corresponding to the sample case, counting the number of correct samples with the verification result consistent with the content of the auditing result, and obtaining a verification rule model when the ratio of the number of the correct samples to the total number of the samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold value. And finally, acquiring a target case waiting for data verification from the case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case. The related data stored in the platform database of the target case is exhausted by utilizing the computing power of the computer, and a verification rule model is generated by comparing the related data with the corresponding indexes, so that the complete analysis of the related data is realized, and the accuracy of the verification result of the target case is further improved.

Further, as a specific implementation of the method shown in fig. 1, an embodiment of the present application provides a data verification apparatus, and as shown in fig. 3, the apparatus includes: a determination module 301, an alignment module 302, a calculation module 303, and a verification module 304.

The determining module 301 is configured to obtain a plurality of sample cases from a historical verification database, and determine a verification result corresponding to each sample case according to a plurality of sample associated data corresponding to each sample case, where the sample associated data is service data generated by a sample client corresponding to the sample case on a platform;

the comparison module 302 is configured to compare the verification result with the audit result corresponding to the sample case, and count the number of correct samples with the verification result consistent with the audit result;

the calculating module 303 is configured to obtain a calibration rule model when a ratio of the number of correct samples to a total number of samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold;

the verification module 304 is configured to obtain a target case waiting for data verification in a case database, input the target case to the verification rule model for data verification, and obtain a target verification result of the target case.

In a specific application scenario, the determining module 301 is configured to, for each sample case in the multiple sample cases, read text type information of the sample case, and extract sample client information and sample audit data corresponding to the sample case from the text type information, where the sample client information is used to indicate a sample client corresponding to the sample case, and the sample audit data is used to indicate a case type corresponding to the sample case; calling an external system connection interface, identifying client tags of all service data in an external service database, determining a plurality of specified client tags with tag contents consistent with the sample client information, and taking a plurality of service data corresponding to the specified client tags as the sample associated data; inquiring a plurality of check rule items corresponding to the sample audit data in a check rule database, and performing data check on the plurality of sample associated data according to the plurality of check rule items to obtain a plurality of check results; in the plurality of verification results, dividing at least one verification result with consistent result content into the same result group to obtain a plurality of result groups; determining a total result ratio corresponding to each result group in the plurality of result groups to obtain a plurality of total result ratios; sequencing the total result ratios, determining a designated total result ratio with the ranking meeting a preset condition, and taking a verification result corresponding to the designated total result ratio as a verification result corresponding to the sample case; for each sample case in the plurality of sample cases, obtaining a plurality of sample associated data corresponding to each sample case, determining a verification result of each sample case, and obtaining a plurality of verification results.

In a specific application scenario, the determining module 301 is configured to determine, for each sample associated data in the plurality of sample associated data, a data category of the sample associated data, and query, in the plurality of check rule items, a first designated check rule item whose item identifier is consistent with the data category; verifying the sample associated data according to a verification index corresponding to the first specified verification rule item, and extracting a verification result hit by the sample associated data as a verification result corresponding to the sample associated data; and verifying each sample associated data in the plurality of sample associated data to obtain a plurality of verification results.

In a specific application scenario, the determining module 301 is configured to, for each result group of the plurality of result groups, identify at least one verification result included in the result group; determining a second specified verification rule item corresponding to each verification result in the at least one verification result, and taking the proportion weight corresponding to the second specified verification rule item as the result proportion corresponding to each verification result to obtain at least one result proportion corresponding to the at least one verification result; adding the at least one result ratio to obtain a total result ratio corresponding to each result group; and calculating the total result ratio corresponding to each result group in the plurality of result groups to obtain the plurality of total result ratios.

In a specific application scenario, the comparison module 302 is configured to identify an audit result corresponding to each sample case in the multiple sample cases, and compare the audit result corresponding to each sample case with a verification result corresponding to each sample case; determining a target sample case with a verification result consistent with an auditing result in the plurality of sample cases, and adding an auditing label for indicating correct content to the target sample case; and counting the number of target sample cases added with the audit tags for indicating the correct contents to obtain the correct sample number.

In a specific application scenario, the apparatus further includes: a statistics module 305.

The statistic module 305 is configured to determine a verification completion time point for obtaining the target verification result, and continuously count a time interval between a current time point and the verification completion time point;

the verification module 304 is configured to, when the time interval reaches a preset time interval, obtain a new target case again in the case database, input the new target case to the verification rule model for data verification, and obtain a target verification result of the new target case.

In a specific application scenario, the apparatus further includes: a presentation module 306, a marking module 307, an extraction module 308, and a storage module 309.

The display module 306 is configured to send the verification result to an audit terminal for displaying, and obtain an uploaded audit result of the audit terminal;

the marking module 307 is configured to, when the audit result indicates that the verification result is changed, extract the audit result and the modification reason uploaded by the audit terminal, mark the sample case information with the audit result, and store the marked sample case information in the historical verification database;

the extracting module 308 is configured to extract a target verification rule item and a modification index from the modification reason, and query a verification rule database for a target verification index corresponding to the target verification rule item;

the storage module 309 is configured to update the target verification index by using the modification index, and store the updated target verification index into the verification rule database.

According to the device provided by the embodiment of the application, a plurality of sample cases are firstly obtained from a historical verification database, and the verification result corresponding to each sample case is determined according to a plurality of sample associated data corresponding to each sample case. And then, comparing the checking result with the auditing result corresponding to the sample case, counting the number of correct samples with the same content of the checking result and the auditing result, and obtaining a checking rule model when the ratio of the number of the correct samples to the total number of the samples of the plurality of sample cases is greater than or equal to a preset accuracy threshold value. And finally, acquiring a target case waiting for data verification in the case database, inputting the target case into the verification rule model for data verification, and obtaining a target verification result of the target case. The related data stored in the platform database of the target case is exhausted by utilizing the computing power of the computer, and a verification rule model is generated by comparing the related data with the corresponding indexes, so that the complete analysis of the related data is realized, and the accuracy of the verification result of the target case is further improved.

It should be noted that other corresponding descriptions of the functional units related to the data verification apparatus provided in the embodiment of the present application may refer to the corresponding descriptions in fig. 1 and fig. 2, and are not described herein again.

In an exemplary embodiment, referring to fig. 4, there is further provided a device, which includes a bus, a processor, a memory and a communication interface, and may further include an input/output interface and a display device, wherein the functional units may communicate with each other through the bus. The memory stores computer programs, and the processor is used for executing the programs stored in the memory and executing the data verification method in the embodiment.

A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the data verification method.

Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by hardware, and can also be implemented by software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.), and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the implementation scenarios of the present application.

Those skilled in the art will appreciate that the figures are merely schematic representations of one preferred implementation scenario and that the blocks or flow diagrams in the figures are not necessarily required to practice the present application.

Those skilled in the art will appreciate that the modules in the devices in the implementation scenario may be distributed in the devices in the implementation scenario according to the description of the implementation scenario, or may be located in one or more devices different from the present implementation scenario with corresponding changes. The modules of the implementation scenario may be combined into one module, or may be further split into a plurality of sub-modules.

The above application serial numbers are for description purposes only and do not represent the superiority or inferiority of the implementation scenarios.

The above disclosure is only a few specific implementation scenarios of the present application, but the present application is not limited thereto, and any variations that can be made by those skilled in the art are intended to fall within the scope of the present application.

Claims

1. A method for data verification, comprising:

obtaining a plurality of sample cases from a historical verification database, and determining a verification result corresponding to each sample case according to a plurality of sample associated data corresponding to each sample case, wherein the sample associated data is service data generated by a sample client corresponding to the sample case on a platform;

2. The method of claim 1, wherein said determining the verification result for each of the plurality of sample cases based on the plurality of sample association data corresponding to said each of the plurality of sample cases comprises:

sequencing the total result ratios, determining a designated total result ratio of which the ranking meets a preset condition, and taking a sub-verification result corresponding to the designated total result ratio as a verification result corresponding to the sample case;

3. The method of claim 2, wherein the performing data verification on the plurality of sample-related data according to the plurality of verification rule items to obtain a plurality of sub-verification results comprises:

verifying the sample associated data according to a verification index corresponding to the first specified verification rule item, and extracting a verification result hit by the sample associated data as a sub-verification result corresponding to the sample associated data;

and verifying each sample associated data in the plurality of sample associated data to obtain a plurality of sub-verification results.

4. The method of claim 2, wherein determining a total outcome proportion for each of the plurality of outcome groups to obtain a plurality of total outcome proportions comprises:

for each result set of the plurality of result sets, identifying at least one sub-verification result contained in the result set;

5. The method according to claim 1, wherein the comparing the plurality of verification results with the audit results corresponding to the plurality of sample cases, and the counting the number of correct samples with the verification results consistent with the audit results in the plurality of verification results comprises:

and counting the number of target sample cases added with the audit tags for indicating the correct contents to obtain the correct sample number.

6. The method according to claim 1, wherein the target case waiting for data verification is obtained from a case database, the target case is input to the verification rule model for data verification, and after a target verification result of the target case is obtained, the method further comprises:

7. The method of claim 1, further comprising:

when the verification result indicates that the verification result is changed, extracting the verification result and a modification reason uploaded by the verification terminal, marking the target case information by using the verification result, and storing the marked target case information into the historical verification database;

8. A data verification apparatus, comprising:

9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.