Search

CN-121981825-A - Business risk identification method and device

CN121981825ACN 121981825 ACN121981825 ACN 121981825ACN-121981825-A

Abstract

The application provides a business risk identification method and a business risk identification device, which can be applied to the technical field of artificial intelligence. The business risk identification method comprises the steps that under the condition that business risk data are received by a model recommending agent, at least one risk identification model is matched for business risks included in the business risk data from a model library, the risk identification model prescribes transaction fields to be checked and check rules for the transaction fields when risk identification is carried out, a clue checking agent obtains the transaction data from a transaction record to be checked according to the transaction fields, the transaction data are input into the risk identification model, and the risk identification model checks the transaction data according to the check rules to obtain a risk identification result.

Inventors

  • NI JINXIAO

Assignees

  • 中国工商银行股份有限公司

Dates

Publication Date
20260505
Application Date
20260120

Claims (10)

  1. 1. A business risk identification method, comprising: Under the condition that the model recommending agent receives the business risk data, matching at least one risk identification model for the business risk included in the business risk data from a model library, wherein the risk identification model prescribes transaction fields needing to be checked when risk identification is carried out and checking rules for the transaction fields; The clue verifying agent acquires transaction data from the transaction record to be checked according to the transaction field, and inputs the transaction data into the risk identification model; And the risk identification model checks the transaction data according to the checking rule to obtain a risk identification result.
  2. 2. The method of claim 1, wherein the model recommending that the agent, upon receiving the business risk data, match at least one risk identification model from a model library for the business risk included in the business risk data comprises: determining at least one service execution step causing service risk according to the service risk; And determining a risk identification model for checking the execution condition of the business execution step from the model library as a risk identification model matched with the business risk.
  3. 3. The method according to claim 2, wherein said determining at least one business execution step that results in a business risk based on said business risk comprises: carrying out semantic analysis on service management specifications of the service, and determining a corresponding relation between a service execution step and service risks; and determining a service execution step which causes service risk according to the corresponding relation.
  4. 4. The method as recited in claim 1, wherein the method further comprises: the risk research judgment agent determines a checking data range according to the business risk data, wherein the checking data range comprises at least one of a data source range and a data time range; And the clue checking agent screens the transaction records to be checked according to the checking data range to obtain screened transaction records to be checked, and acquires transaction data input into the risk identification model from the screened transaction records to be checked.
  5. 5. The method of any one of claims 1-4, further comprising: the risk research judgment agent acquires at least one of supervision punishment data, public opinion data and financial policy data, performs semantic analysis, and determines business risk data.
  6. 6. The method according to any one of claims 1 to 4, wherein the checking rule includes at least one judgment condition corresponding to a business risk, and the risk identification model checks the transaction data according to the checking rule to obtain a risk identification result includes: And under the condition that the transaction data accords with at least one judging condition corresponding to the business risk, determining that the risk identification result is that the transaction record has the business risk.
  7. 7. A business risk identification device, comprising: The model recommending module is used for calling the model recommending agent to match at least one risk identifying model for the business risk included in the business risk data from the model library under the condition of receiving the business risk data, wherein the risk identifying model prescribes transaction fields needing to be checked when risk identification is carried out and checking rules for the transaction fields; The clue checking module is used for calling a clue checking agent to acquire transaction data from the transaction records to be checked according to the transaction field and inputting the transaction data into the risk identification model; And the risk identification result determining module is used for calling a risk identification model to check the transaction data according to the checking rule to obtain a risk identification result.
  8. 8. An electronic device, comprising: One or more processors; a memory for storing one or more computer programs, Characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1-6.
  9. 9. A computer-readable storage medium, on which a computer program or instructions is stored, which, when executed by a processor, carries out the steps of the method according to any one of claims 1-6.
  10. 10. A computer program product comprising a computer program or instructions which, when executed by a processor, implement the steps of the method according to any one of claims 1-6.

Description

Business risk identification method and device Technical Field The application relates to the field of artificial intelligence, in particular to a business risk identification method and device. Background In a financial business scenario, it is often necessary to determine whether a customer transaction has one or more business risks, and if each transaction record has a business risk, it is necessary to manually check each transaction record to determine whether the customer transaction has a business risk, so that the efficiency of identifying the business risk is low. For this reason, a new business risk identification method is required. Disclosure of Invention In view of the above problems, the present application provides a business risk identification method and apparatus. According to the first aspect of the application, a business risk identification method is provided, which is characterized by comprising the steps that under the condition that business risk data are received by a model recommending agent, at least one risk identification model is matched for business risks included in the business risk data from a model library, the risk identification model prescribes a transaction field to be checked and a checking rule for the transaction field when risk identification is carried out, a clue checking agent acquires transaction data from a transaction record to be checked according to the transaction field and inputs the transaction data into the risk identification model, and the risk identification model checks the transaction data according to the checking rule to obtain a risk identification result. According to the embodiment of the application, the model recommending agent is used for matching at least one risk identification model for the business risk included in the business risk data in the model library under the condition of receiving the business risk data, wherein the method comprises the steps of determining at least one business execution step causing the business risk according to the business risk, and determining a risk identification model for checking the execution condition of the business execution step from the model library as a risk identification model matched with the business risk. According to the embodiment of the application, the service execution step for determining at least one service causing the service risk according to the service risk comprises the steps of carrying out semantic analysis on a service management specification of the service, determining the corresponding relation between the service execution step and the service risk, and determining the service execution step causing the service risk according to the corresponding relation. The embodiment of the application further comprises a model recommending agent determining a checking data range according to the business risk data, wherein the checking data range comprises at least one of a data source range and a data time range, and the clue checking agent screening transaction records to be checked according to the checking data range to obtain screened transaction records to be checked and obtaining transaction data input into the risk identification model from the screened transaction records to be checked. According to the embodiment of the application, the risk research judgment agent acquires at least one of supervision punishment data, public opinion data and financial policy data and performs semantic analysis to determine business risk data. According to the embodiment of the application, the method comprises the steps that the checking rule comprises at least one judging condition corresponding to the business risk, and the risk identification model checks the transaction data according to the checking rule to obtain a risk identification result, wherein the risk identification result is determined to be that the transaction record has the business risk under the condition that the transaction data accords with the at least one judging condition corresponding to the business risk. The application provides a business risk identification device which is characterized by comprising a model recommendation module, a clue verification module and a risk identification result determination module, wherein the model recommendation module is used for calling a model recommendation agent to match business risks contained in business risk data from a model library with at least one risk identification model under the condition that the business risk data are received, the risk identification model prescribes a transaction field to be checked and a check rule for the transaction field when risk identification is carried out, the clue verification module is used for calling the clue verification agent to acquire transaction data from a transaction record to be checked according to the transaction field and input the transaction data into the risk identification model, and the risk identification res