Effective financial institution intelligent inspection system and method

Abstract

The invention discloses an effective financial institution intelligent inspection system and method, and relates to the field of finance. The invention discloses an effective financial institution intelligent inspection system. A screening result of a risk monitoring model is identified as evidence information with risks, and the evidence information is brought into an early warning clue library; identifying whether the early warning clue is an inspection clue according to the transaction evidence chain; obtaining inspection key points according to management regulations of the financial institution; according to the risk scores corresponding to the question library, forming a risk level library of the inspection questions of the inspected object; creating an inspection task according to the risk level library and the inspector management library of the inspected object; and automatically distributing the inspection content of the inspection task to inspectors for identification. According to the invention, intelligent inspection can be realized for different inspection business scenes and inspection contents of the financial institution.

Description

Technical Field

[0001] The present invention relates to the financial field, and in particular to an effective intelligent inspection system and method for financial institutions. Background Art

[0002] On-site inspections of financial institutions mainly include daily compliance and internal control inspections, operational management inspections, and audit on-site inspections. The inspection methods mainly include routine inspections, temporary inspections, and external regulatory inspections. At present, traditional on-site inspections of financial institutions provide fixed inspection templates to create inspection tasks based on the type of inspection items. Although they can meet the needs of on-site inspections, they are not very targeted, the inspection results are not comprehensive, and they also cause a waste of inspection resources. With the development of information technology, the intelligent and digital transformation of financial institutions has accelerated, and various business systems and data systems are also constantly strengthening digital construction. Financial institutions are also constantly innovating in on-site inspection methods to cope with the development of digital business.

[0003] With the digital transformation of financial institutions, the scope and volume of business data are also growing rapidly, and business transaction scenarios and management scenarios have undergone significant changes. Relying on existing inspection methods, the inspection content is not focused, the inspection personnel are relatively rigidly assigned, and the inspection results are not comprehensive and targeted. The shortcomings of existing on-site inspection technologies of financial institutions are mainly in the following aspects: 1) There is a lack of intelligent risk assessment system. The risk assessment and classification of inspection objects cannot be carried out through the risk assessment system. Targeted inspection plans are mostly solidified inspection contents or partially introduced digital information contents. The inspection plans and inspection contents formed are basically the same templates, which are not very targeted.

[0004] 2) Failure to dynamically manage and intelligently analyze inspectors, and instead conducting inspections based on fixed daily staffing arrangements, inspection resources were not accurately allocated, which had a direct impact on the effectiveness of the inspections.

[0005] 3) After completing the inspection, many financial institutions fail to store and track the inspection issues and conduct subsequent re-inspections, thereby forming a complete closed-loop inspection plan.

[0006] At present, there is a need for a universal intelligent inspection system and method that can propose a universal intelligent inspection system and method for different inspection business scenarios and inspection contents of financial institutions. Summary of the invention

[0007] The technical problem to be solved by the present invention is how to realize intelligent inspection for different inspection business scenarios and inspection contents of financial institutions. The purpose is to provide an effective intelligent inspection system and method for financial institutions, which solves the problem that the prior art lacks an intelligent risk assessment system, fails to dynamically manage and push intelligent analysis to inspectors, and forms a complete closed-loop inspection solution.

[0008] The present invention is achieved through the following technical solutions: An effective intelligent inspection system for financial institutions, comprising: The risk monitoring model provides early warnings on the transaction behaviors of financial institutions and customers based on the business data of financial institutions; The early warning clue monitoring module includes the risk evidence information identified by the above risk monitoring model into the early warning clue database; The inspection clue collection module forms a transaction evidence chain based on the transaction behavior of the financial institution and the customer's transaction behavior; manually identifies the warning clues in the warning clue library based on the transaction evidence chain, and uses them as inspection clues when they are identified as risks; The inspection key points collection module obtains the inspection key points according to the management regulations of the financial institution; the inspection clues and the inspection key points are formed into an inspection content library; The object risk assessment module forms a question library of the inspection contents of the inspected object after inspecting the inspected object based on the inspection contents of the inspection content library; The clue intelligent inspection module calculates the risk level of the inspected object according to the risk score corresponding to the question library, and uses the question library and risk level to form a risk level library of the inspection questions of the inspected object; Inspector management module, which identifies the identity of inspectors and forms an inspector management database; The intelligent inspection task module creates inspection tasks based on the risk level library of the inspected object and the inspector management library; automatically assigns the inspection content of the inspection task to the inspector; when the inspector identifies it as risky, the corresponding inspection content is retained, otherwise, the corresponding inspection content in the inspection content library is deleted.

[0009] The above-mentioned financial institution business data includes transaction flow data, customer data and account data; the above-mentioned risk monitoring model, based on the financial institution business data, issues early warnings on the financial institution's transaction behavior and customer transaction behavior, including: when constructing the above-mentioned risk monitoring model, it is defined as issuing early warnings on the financial institution's transaction behavior from the dimensions of the financial institution's compliance risk characteristics and operational risk characteristics, and issuing early warnings on customer transaction behavior from the dimensions of customer transaction behavior characteristics and customer attribute characteristics; the above-mentioned evidence information includes the compliance risk characteristics and operational risk characteristics of the financial institution, as well as any one or more of the risk subject's transaction behavior characteristics and customer attribute characteristics.

[0010] The above-mentioned transaction behavior characteristics include any one or more of transaction amount, transaction status, number of transactions and transaction time; the above-mentioned customer attribute characteristics include any one or more of main customers, core customers, ordinary customers, corporate customers, non-bank customers and non-bank corporate customers.

[0011] The above-mentioned early warning clue monitoring module includes a real-time clue monitoring module and a flow monitoring module; the above-mentioned real-time clue monitoring module is used to give real-time early warning to the currently selected financial institutions and customers; the above-mentioned flow monitoring module is used to give early warning to the currently selected financial institutions and customers according to time intervals.

[0012] The above screening results identified as risky include: large amounts of cash transactions in personal accounts, frequent cash deposits with large accumulated amounts, deposits and concentrated insurance before and after holidays, huge transactions in personal accounts in a short period of time and huge account balances, and any one or more of the following information: cash deposit and withdrawal transactions by personal customers at multiple institutions or outlets in one day.

[0013] The above screening results identified as risky include: large amounts of cash transactions in personal accounts, frequent cash deposits with large accumulated amounts, deposits and concentrated insurance before and after holidays, huge transactions in personal accounts in a short period of time and huge account balances, and any one or more of the following information: cash deposit and withdrawal transactions by personal customers at multiple institutions or outlets in one day.

[0014] When creating inspection tasks based on the risk level database of the inspected object and the inspection personnel management database, it includes the method of creating inspection tasks by random inspection scheme; The above random inspection program includes: According to the input number of random inspection objects, the inspection objects are selected in order from the largest to the smallest period when the inspection objects last completed the inspection task; the inspection objects whose risk level of the last inspection task completed is the same as the risk level of the inspected objects are selected; and the inspection points and inspection clues corresponding to the inspection problems of the inspected objects are collected from the inspection content library according to the risk level library.

[0015] An effective intelligent inspection method for financial institutions includes: The risk monitoring model issues early warnings on the transaction behaviors of financial institutions and customers based on the business data of financial institutions; The risk monitoring model's screening results are identified as evidence of risk and are included in the early warning clue database; A transaction evidence chain is formed based on the transaction behavior of financial institutions and customer transaction behavior; the early warning clues in the early warning clue library are manually identified based on the transaction evidence chain, and when identified as risks, they are used as inspection clues; Obtain inspection key points according to the management regulations of financial institutions; form the inspection content library with the above inspection clues and the above inspection key points; After inspecting the inspected object based on the inspection contents in the inspection content library, a question library of the inspection contents of the inspected object is formed; According to the risk scores corresponding to the question library, the risk level of the inspected object is calculated, and the risk level library of the inspection questions of the inspected object is formed by using the question library and risk level; Identify the identities of inspectors and form an inspector management database; Create inspection tasks based on the risk level library of the inspected object and the inspector management library; automatically assign the inspection content of the inspection task to the inspector; when the inspector identifies it as having risks, retain the corresponding inspection content, otherwise delete the corresponding inspection content in the inspection content library.

[0016] An electronic device comprises a memory, a processor and a computer program running on the processor, wherein the processor implements an effective intelligent inspection system for financial institutions as described above when executing the computer program.

[0017] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements an effective intelligent inspection system for financial institutions as described above.

[0018] Compared with the prior art, the present invention has the following advantages and beneficial effects: The effective intelligent inspection method for financial institutions in this application is supported by artificial intelligence and big data, and proposes a general intelligent inspection method for different inspection business scenarios and inspection contents of financial institutions. This method identifies the transaction behaviors of financial institutions and customers through big data risk monitoring technology, and analyzes the inspection clue library in a targeted manner. At the same time, it defines the inspection key points library in combination with management regulations. The inspection personnel are managed statically and dynamically, and matching inspection personnel are autonomously identified based on personnel management data. The system automatically creates inspection tasks for business personnel to conduct inspections. It enhances the in-depth understanding and risk identification capabilities of complex financial businesses, and improves the quality and efficiency of risk inspections. The present invention can conduct intelligent inspections for different inspection business scenarios and inspection contents of financial institutions, solving the problem that the prior art lacks an intelligent risk assessment system, fails to dynamically manage and intelligently analyze inspection personnel, and forms a complete closed-loop inspection solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 This is a flow chart of an intelligent inspection system for financial institutions that is effective in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention. Example

[0021] like Figure 1 As shown, the embodiment of the present application provides an effective intelligent inspection system for financial institutions, including: The risk monitoring model provides early warnings on the transaction behaviors of financial institutions and customers based on the business data of financial institutions; The early warning clue monitoring module includes the risk evidence information identified by the above risk monitoring model into the early warning clue database; The inspection clue collection module forms a transaction evidence chain based on the transaction behavior of the financial institution and the customer's transaction behavior; manually identifies the warning clues in the warning clue library based on the transaction evidence chain, and uses them as inspection clues when they are identified as risks; The inspection key points collection module obtains the inspection key points according to the management regulations of the financial institution; the inspection clues and the inspection key points are formed into an inspection content library; The object risk assessment module forms a question library of the inspection contents of the inspected object after inspecting the inspected object based on the inspection contents of the inspection content library; The clue intelligent inspection module calculates the risk level of the inspected object according to the risk score corresponding to the question library, and uses the question library and risk level to form a risk level library of the inspection questions of the inspected object; Inspector management module, which identifies the identity of inspectors and forms an inspector management database; The intelligent inspection task module creates inspection tasks based on the risk level library of the inspected object and the inspector management library; automatically assigns the inspection content of the inspection task to the inspector; when the inspector identifies it as risky, the corresponding inspection content is retained, otherwise, the corresponding inspection content in the inspection content library is deleted.

[0022] The above-mentioned financial institution business data includes transaction flow data, customer data and account data; the above-mentioned risk monitoring model, based on the financial institution business data, issues early warnings on the financial institution's transaction behavior and customer transaction behavior, including: when constructing the above-mentioned risk monitoring model, it is defined as issuing early warnings on the financial institution's transaction behavior from the dimensions of the financial institution's compliance risk characteristics and operational risk characteristics, and issuing early warnings on customer transaction behavior from the dimensions of customer transaction behavior characteristics and customer attribute characteristics; the above-mentioned evidence information includes the compliance risk characteristics and operational risk characteristics of the financial institution, as well as any one or more of the risk subject's transaction behavior characteristics and customer attribute characteristics.

[0023] The above-mentioned transaction behavior characteristics include any one or more of transaction amount, transaction status, number of transactions and transaction time; the above-mentioned customer attribute characteristics include any one or more of main customers, core customers, ordinary customers, corporate customers, non-bank customers and non-bank corporate customers.

[0024] The above-mentioned early warning clue monitoring module includes a real-time clue monitoring module and a flow monitoring module; the above-mentioned real-time clue monitoring module is used to give real-time early warning to the currently selected financial institutions and customers; the above-mentioned flow monitoring module is used to give early warning to the currently selected financial institutions and customers according to time intervals.

[0025] The above screening results identified as risky include: large amounts of cash transactions in personal accounts, frequent cash deposits with large accumulated amounts, deposits and concentrated insurance before and after holidays, huge transactions in personal accounts in a short period of time and huge account balances, and any one or more of the following information: cash deposit and withdrawal transactions by personal customers at multiple institutions or outlets in one day.

[0026] When creating inspection tasks based on the risk level library of the inspected object and the inspector management library, it includes a method of creating inspection tasks by specifying the inspection plan; The above-mentioned designated inspection plan includes: selecting multiple inspected objects with the highest risk level ranking to generate an inspection plan; selecting the above-mentioned inspection clues that need to be inspected from the above-mentioned inspection point library according to the identity of the above-mentioned inspection object; selecting the above-mentioned inspection points that need to be inspected from the above-mentioned inspection point library according to the risk issues when the above-mentioned inspected objects are evaluated for risk levels; and assigning inspection tasks to inspectors in the above-mentioned inspection personnel management library according to the implementation cycle of the inspection project and the content of the inspection tasks.

[0027] When creating inspection tasks based on the risk level database of the inspected object and the inspection personnel management database, it includes the method of creating inspection tasks by random inspection scheme; The above random inspection program includes: According to the input number of random inspection objects, the inspection objects are selected in order from the largest to the smallest period when the inspection objects last completed the inspection task; the inspection objects whose risk level of the last inspection task completed is the same as the risk level of the inspected objects are selected; and the inspection points and inspection clues corresponding to the inspection problems of the inspected objects are collected from the inspection content library according to the risk level library.

[0028] Among them, when monitoring the transaction behaviors of financial institutions and customers through risk monitoring models, screen for evidence information that meets transaction risks. For example, when the screening result is the risk of employees taking out loans to purchase financial products, the evidence information includes: the transaction type is the purchase of financial products, the source of funds is the loan issued, and the customer who purchased the financial product is an employee.

[0029] The embodiment of the present invention mainly includes inspection clue collection, object risk assessment, inspection personnel management, manual inspection tasks and clue intelligent inspection modules, so as to realize the collection of inspection content, risk level assessment of the inspected object after the inspection content collection is completed, form a risk level library of the inspection object, and then dynamically identify the inspectors to form an inspection personnel management library. After obtaining the inspection object risk level and the inspection personnel library, the system intelligently creates inspection tasks, automatically assigns inspection points and inspection clues according to the inspection objects recommended by the system, and the inspectors conduct inspections according to the assigned tasks. The inspection results generated during the inspection may include inspection questions and inspection reports, so as to iteratively optimize the inspection content.

[0030] A complete inspection content system has been formed, which is mainly divided into two parts: Identify the inspection clues of the inspected object. The inspection clues are based on business data and match and monitor the business data of financial institutions through the characteristics of risk monitoring. According to the rules of risk definition, artificial intelligence technology is used to identify early warning clues. With the inspection object as the center, the transaction content as the main line, and the risk clue scenario as the basis, the compliance risk characteristics and operational risk characteristics of financial institutions, as well as the transaction behavior characteristics and customer attribute characteristics of risk subjects are identified. In addition, according to the flow of transaction funds between financial institutions and customers, a transaction evidence chain of multiple transaction objects is formed, thereby providing a basis for early warning clues to identify manual risks. Among them, the compliance risk of financial institutions refers to the risk of internal control compliance management of institutions, and the operational risk of financial institutions refers to the risk caused by illegal operations of employees. The transaction evidence chain is a fund flow diagram of multiple transaction objects, including the transaction characteristics of each transaction (transaction object and transaction behavior), for inspectors to further judge and identify risks. The transaction evidence chain displays each transaction based on the internal control compliance management of financial institutions and employee operations, customer attributes and customer transaction behaviors. The inspection points obtained according to the management regulations of financial institutions include multiple business dimensions, such as credit, core, and related businesses conducted by peers. The transaction behavior characteristics of customers can also include transaction channels, transaction locations, etc. For example, when the operational risk of a financial institution is "abnormal time of personal online banking non-financial transactions", it specifically refers to the teller handling non-financial transactions at the personal online banking counter during non-working hours. Non-financial transactions include: account opening, card number association, password change, certificate update, limit change, etc.

[0031] Obtaining inspection points according to the management regulations of financial institutions is to build a comprehensive inspection point library from the internal management system and external supervision requirements of financial institutions. The internal management system is the management requirements of relevant departments, such as credit compliance management, accounting counter compliance management, online banking, mobile phone transactions and other business compliance management. If there is no compliance, there will be risks. The inspection points are defined on this basis. External supervision is mainly the risk compliance management requirements of relevant departments for financial institutions, such as customer management, account management, etc.; it also involves the management of accounts involved in the case and customers of electronic fraud. Inspection clues library for financial institutions, such as "employee loans to purchase financial management", "non-employee loan funds transferred to employee accounts", "related enterprise loan issuance", "large transfers from public accounts to employee accounts", etc. Inspection clues library for customers, such as "large cash transactions in personal accounts", "frequent cash deposits and large cumulative amounts", "concentrated deposits and insurance before and after holidays", "huge transactions in private accounts within a month, large account balances", "private customers make cash deposit and withdrawal transactions at multiple machines / outlets within a day", etc. Business personnel will then combine their business experience to determine whether there are risks in the transaction evidence chain and whether to include the warning clues in the inspection clue library.

[0032] The key points for checking different project categories are shown in Table 1 below:

[0033]

[0034] Table 1 Optionally, based on the number of problems of the inspection object, the risk scores of the problem terms corresponding to the problem database are counted, and the risk values ​​of the inspection objects are ranked from the annual, quarterly and monthly dimensions. The risk level and score of the problem terms can be customized according to the risk management department, such as the compliance department and the audit department, based on the risk of the loss caused to the relevant departments.

[0035] Optionally, the inspector management database includes two parts: actual dynamics and planned dynamics. The actual dynamics include automatic statistical information such as the number of inspection projects, the number of times as the inspection team leader, the work plan for the current year, the role played in the last project, and the attendance of personnel; the planned dynamics include information manually recorded on personnel leave and training, so as to facilitate the arrangement of idle personnel. Optionally, based on the experience of idle inspectors, the inspectors matching the corresponding inspection projects are identified and automatically assigned to the inspection projects. The inspection tasks mainly include: economic responsibility inspection, comprehensive business inspection, special business inspection, internal control inspection, accountability inspection and other types. The arranged personnel matching the inspection project have experience in participating in the same type of projects and taking on the inspection content, and can be automatically assigned based on the inspection results and personnel with higher assessment scores.

[0036] The embodiment of this application provides two schemes: system-specified and system-random, which realizes the autonomous creation of intelligent inspection tasks combined with the current situation of financial institutions. The system-specified inspection scheme is the risk level ranking of the inspection object evaluated by the system. When initiating the system-specified inspection scheme, the system automatically assigns the inspection scheme to the top 10 inspection objects ranked by the assessed risk level (this parameter can be configured by the system). The generation logic of the inspection scheme is as follows: Inspection clue collection: Based on the inspector, the system automatically selects the inspection clues that the inspector needs to check and brings them into the inspection project as inspection tasks.

[0037] Inspection key points collection: Inspection key points are based on the problems involved in the inspection object, the problem terms classified by the problem, and the corresponding business classification according to the classification. Finally, specific inspection key points can be collected according to the business classification and included in the inspection project tasks. For example, if the problem term involved in the problem database is "the handover of bill vouchers is not registered and causes loss", then the system extracts the inspection key points such as "whether the storage of paper bills has approval for entry and exit, whether it is strictly handed over by two people, and whether it is regularly inspected" through the problem term business classification.

[0038] Assignment of inspectors: Assignment of inspectors is done through the dynamic management database of inspectors. According to the implementation cycle of the inspection project and the inspection task content of the inspection project, the system pushes the person in charge of the inspection project, the team leader and the project team members who meet the requirements. The appropriate personnel are pushed to the project as the preliminary selection results. The person in charge of the inspection department will check and confirm the personnel pushed by the system. After the check is completed, the project personnel will be assigned. After the above three parts are completed, the project tasks specified by the system are created, and the project implementation inspection phase will begin after the project is started.

[0039] The inspection plan designated by the system is mainly aimed at inspection objects with higher risk assessment levels. In order to fully cover the risks of the inspected institutions, a system random inspection plan is also proposed. This plan is also a good supplement to the system designated plan. System random inspection plan: According to the inspection needs of the inspection department, limited resources can be used to complete key inspection work. According to the inspection time range, the resource situation of the inspectors can be screened in the inspector management database. According to the resource situation, the random number of inspection objects, the number of inspection points and clues, and the time period of the inspection are entered in the random inspection plan, and the system generates the inspection task. The generation logic of the inspection plan is as follows: Inspection object collection: The system filters out objects that have recently completed inspection tasks based on the number of random inspection objects entered, and collects corresponding inspection objects based on the length of time since the inspection object last completed the inspection task and the risk level.

[0040] Collection of inspection points and inspection clues: The number of inspection points and inspection clues is collected after the inspection object is collected, based on the inspection problem risk level of the object.

[0041] 1) The risk level of the inspection object is based on the number of questions corresponding to the inspection object. Each question has a corresponding question entry, and each question entry has a corresponding risk score. Finally, the inspection object risk level is graded according to the accumulated risk score.

[0042] 2) The registered problem has a source. If the problem is found during on-site inspection, the business corresponding to the problem entry will have a corresponding inspection point. If the source of the problem is a clue from the risk model warning, the corresponding inspection clue can be obtained based on the problem. For example, if the problem of the inspection object involves interbank deposit accounts, the relevant inspection points need to be included in the inspection task. At the same time, if the inspection object has the problem of "frequent capital business exceeding the authorized authority in a short period of time" in the warning problem, then the inspection clue also needs to be included in the inspection task.

[0043] The inspectors check the currently available personnel in the dynamic personnel inspection database, and the system automatically gives the preliminary candidates. The person in charge of the inspection department will conduct a re-selection and confirmation based on the preliminary candidates pushed by the system. After the re-selection is completed, the project personnel are allocated.

[0044] After the system-specified inspection plan and system random inspection plan tasks are created, the project implementation inspection stage is entered. This stage is to check the content of the inspection task, register inspection questions and working papers according to the inspection results, and submit them to the inspection team leader or inspection project leader for review after the inspection is completed. The reviewed inspection questions and working papers form the inspection task results library. The following is a detailed introduction to the inspection task execution process and cases: The inspectors carry out the inspection task execution: After the project is launched, the inspectors will check according to the inspection points generated by the system during the inspection process, and simultaneously verify and confirm the information in the inspection clue library. After the task is completed, if there are any problems, they can be registered in the problem library according to the inspection results, and the inspection work draft will be written simultaneously. For example: the inspectors will check the inspection points of "payroll business on behalf of others", verify them one by one from the internal compliance management point of payroll on behalf of others, and at the same time check the risk clue information of "cash business in the payroll account account" involving the payroll warning clue on behalf of others in the clue library, which needs to be checked as evidence information at the same time. After the verification is completed, the registration problem that does exist is registered, and the entry of the problem is selected as "the funds are not handled according to regulations when handling the payroll business on behalf of others" to classify the problem. Finally, the above inspection contents are summarized in the work draft to complete the inspection of the current task.

[0045] Inspection task review: The working papers and questions submitted by the project inspectors are reviewed by the project team leader or project manager and entered into the inspection project result database. If the review passes, the inspection point is registered as a positive mark, and if the review fails, it is recorded as a negative mark. The weight of the inspection point with a positive mark is -1, and the weight of the inspection point with a negative mark is +1. This can be used as a statistical inspection object to count the inspection point frequency, and provide data support for the subsequent inspection and push of inspection points.

[0046] The results of the task review are entered into the result database of this inspection for inspection and analysis by the project inspection leader.

[0047] The problems registered by the inspection team members are summarized, duplicate problems are eliminated, and the classification of the terms to which the problems belong, as well as the corresponding risk levels and scores are integrated and adjusted. After the adjustment is completed, it is published to the respective inspected objects for subsequent problem rectification, and it is also used as a registration point for subsequent re-inspection. The project leader is responsible for summarizing the inspection content and writing the inspection report based on the working papers after the inspection tasks completed by the inspection team members.

[0048] In summary, the embodiments of the present application provide an effective intelligent inspection system and method for financial institutions: It has realized the assessment of the risk level of the inspection object and the dynamic management of the inspectors. It has innovated the traditional on-site inspection methods and also used the current big data analysis technology and artificial intelligence learning and recognition methods to improve the quality and efficiency of on-site inspections. The auxiliary inspectors can complete the inspection tasks more quickly, accurately and efficiently, effectively discover compliance problems, and prevent and resolve financial risks. It has realized the creation of precise inspection tasks. By conducting risk assessment and risk level ranking of the inspection objects, it is more accurate and targeted in the process of creating inspection tasks; it can accurately allocate appropriate inspectors and dynamically manage inspectors through intelligent analysis; the rich inspection content library can realize the combination of quantitative business risk warning clues and on-site inspection points, and the inspection content covers a wider range, improving the accuracy and effectiveness of the inspection.

[0049] During the creation of the inspection project, risk issues of the inspection object are collected from four dimensions: off-site risk monitoring, on-site inspection, external inspection, and assessment of the inspection object. At the same time, statistics are collected based on the problem terms associated with the issues and the risk scores corresponding to the problem terms. Finally, the institutions are graded and ranked based on the risk scores. During the creation of the inspection task, the inspection clues and key points of the institution are extracted based on the problem associations, and the clues and key points that have been inspected recently are automatically excluded.

[0050] Inspection tasks are accurately assigned. Through dynamic personnel management combined with system analysis, more suitable personnel can be identified during project inspections, rather than temporary or fixed pattern assignments. Therefore, the quality of inspectors and inspection results are better and more professional.

[0051] The inspection content library is complete and rich. It is able to extract inspection evidence through the inspection clue library, making the inspection more targeted. Through the coverage of the inspection key points library, it can more completely cover the content that needs to be inspected, dig deep into potential problems, and improve the accuracy and effectiveness of on-site inspections.

[0052] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An effective intelligent inspection system for financial institutions, characterized in that: include: The risk monitoring model provides early warnings on the transaction behaviors of financial institutions and customers based on the business data of financial institutions; An early warning clue monitoring module is used to identify the risk evidence information identified by the risk monitoring model as risky and to include it in an early warning clue database; The inspection clue collection module forms a transaction evidence chain based on the transaction behavior of the financial institution and the transaction behavior of the customer; manually identifies the warning clues in the warning clue library according to the transaction evidence chain, and uses them as inspection clues when they are identified as risks; An inspection key point collection module obtains inspection key points according to the management regulations of the financial institution; and forms an inspection content library with the inspection clues and the inspection key points; The object risk assessment module forms a question library of the inspection contents of the inspected object after inspecting the inspected object based on the inspection contents of the inspection content library; The clue intelligent inspection module calculates the risk level of the inspected object according to the risk score corresponding to the question library, and uses the question library and risk level to form a risk level library of the inspection questions of the inspected object; Inspector management module, which identifies the identity of inspectors and forms an inspector management database; The intelligent inspection task module creates inspection tasks based on the risk level library of the inspected object and the inspector management library; automatically assigns the inspection content of the inspection task to the inspector; when the inspector identifies it as risky, the corresponding inspection content is retained, otherwise, the corresponding inspection content in the inspection content library is deleted.

2. An effective intelligent inspection system for financial institutions according to claim 1, characterized in that: The financial institution business data includes transaction flow data, customer data and account data; The risk monitoring model provides early warnings for the transaction behaviors of financial institutions and customers based on the business data of financial institutions, including: The risk monitoring model is defined as providing early warnings for the transaction behaviors of financial institutions from the dimensions of compliance risk characteristics and operational risk characteristics of financial institutions, and providing early warnings for the transaction behaviors of customers from the dimensions of customer transaction behavior characteristics and customer attribute characteristics; The evidence information includes the compliance risk characteristics and operational risk characteristics of the financial institution, and any one or more of the transaction behavior characteristics and customer attribute characteristics of the risk subject.

3. An effective intelligent inspection system for financial institutions according to claim 2, characterized in that: The transaction behavior characteristics include any one or more of transaction amount, transaction status, number of transactions and transaction time; the customer attribute characteristics include any one or more of main customers, core customers, ordinary customers, corporate customers, non-bank customers and non-bank corporate customers.

4. An effective intelligent inspection system for financial institutions according to claim 1, characterized in that: The early warning clue monitoring module includes a real-time clue monitoring module and a flow monitoring module; the real-time clue monitoring module is used to issue real-time early warnings to the currently selected financial institutions and customers; the flow monitoring module is used to issue early warnings to the currently selected financial institutions and customers according to time intervals.

5. An effective intelligent inspection system for financial institutions according to claim 2, characterized in that: The screening results identified as risky include: a large number of cash transactions in personal accounts, frequent cash deposits with large accumulated amounts, deposits and concentrated insurance before and after holidays, huge transaction scales and huge account balances in personal accounts in a short period of time, and any one or more of the following information: cash deposit and withdrawal transactions by personal customers at multiple institutions or outlets in one day.

6. An effective intelligent inspection system for financial institutions according to claim 1, characterized in that: When creating the inspection task according to the risk level library of the inspected object and the inspection personnel management library, it includes a method of specifying the inspection plan to create the inspection task; The designated inspection plan includes: selecting multiple inspected objects with the highest risk level ranking to generate an inspection plan; selecting the inspection clues to be inspected from the inspection point library according to the identity of the inspected object; selecting the inspection points to be inspected from the inspection point library according to the risk issues when the inspected object is evaluated for risk level; and assigning the inspection tasks to the inspectors in the inspection personnel management library according to the implementation cycle of the inspection project and the content of the inspection tasks.

7. An effective intelligent inspection system for financial institutions according to claim 1, characterized in that: When creating inspection tasks according to the risk level library of the inspected object and the inspection personnel management library, it includes a method of creating inspection tasks by random inspection schemes; The random inspection program includes: According to the input number of random inspection objects, the inspection objects are selected in order from large to small according to the period when the inspection objects last completed the inspection task; the inspection objects whose risk level of the last inspection task completed is the same as the risk level of the inspected objects are selected; and the inspection points and inspection clues corresponding to the inspection problems of the inspected objects are collected from the inspection content library according to the risk level library.

8. An effective intelligent inspection method for financial institutions, characterized in that: include: The risk monitoring model issues early warnings on the transaction behaviors of financial institutions and customers based on the business data of financial institutions; The risk monitoring model identifies the risk evidence information as a result of the screening and includes it in the early warning clue database; A transaction evidence chain is formed based on the transaction behavior of financial institutions and the transaction behavior of customers; the early warning clues in the early warning clue library are manually identified based on the transaction evidence chain, and when identified as risks, they are used as inspection clues; Obtaining inspection key points according to the management regulations of the financial institution; forming an inspection content library with the inspection clues and the inspection key points; After inspecting the inspected object based on the inspection contents in the inspection content library, a question library of the inspection contents of the inspected object is formed; According to the risk scores corresponding to the question library, the risk level of the inspected object is calculated, and the risk level library of the inspection questions of the inspected object is formed by using the question library and risk level; Identify the identities of inspectors and form an inspector management database; Create inspection tasks based on the risk level library of the inspected object and the inspector management library; automatically assign the inspection content of the inspection task to the inspector; when the inspector identifies it as having risks, retain the corresponding inspection content, otherwise delete the corresponding inspection content in the inspection content library.

9. An electronic device comprising a memory, a processor, and a computer program running on the processor, characterized in that: When the processor executes the computer program, an effective intelligent checking system for financial institutions as claimed in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, an effective intelligent checking system for financial institutions as claimed in any one of claims 1 to 7 is implemented.

Images