CN117151857A - Approval risk screening method, device, equipment and storage medium thereof - Google Patents

Abstract

The embodiment of the application belongs to the technical field of financial science and technology, and relates to an approval risk screening method, an approval risk screening device, approval risk screening equipment and a storage medium thereof, wherein the method comprises the steps of obtaining approved historical approval information; screening all the historical approval information corresponding to the same request main body, and carrying out comprehensive analysis; performing risk screening on the historical request behaviors of different request subjects and the corresponding historical approval results, and screening out risk request subjects and risk approval subjects; and sending the request subject identification of the risk request subject and the approval subject identification of the risk approval subject to a preset risk control end for risk control processing. In the loan application and credit card application business of the finance company, bidirectional control is performed by analyzing the behaviors of the request main body and the behaviors of the auditing main body, so that the illegal approval behaviors of the finance business are reduced as much as possible, and the risk control degree of the finance business is improved.

Description

Approval risk screening method, device, equipment and storage medium thereof

Technical Field

The application relates to the technical field of financial science and technology, and is applied to a financial business approval scene, in particular to an approval risk screening method, an approval risk screening device, approval risk screening equipment and a storage medium thereof.

Background

With the rapid development of the financial industry, more and more financial companies are involved in approval or audit services, such as bank loan service and credit card opening service, and often, the financial companies can finally determine whether to pay the credit or send down a card to the applicant by checking the applicant layer by layer.

In the current finance company, through manual mode review, though both have relatively fixed approval chain and application requirement standard, but applicant's application reason is different, and information identification etc. of its superior leader and cross-level leader all need go on the verification through the manual work that utilizes a large amount of time in a large amount, in the manual verification process, once sign the volume and surpass a certain amount, on the basis of a small amount of manpower, the people can't avoid appearing the mental of lackluster, lazy, the result quality of manual audit becomes very uncontrollable, be unfavorable for the rigor of this work, prior art still lacks the scheme of discerning risk application main part and risk audit main part simultaneously, can't carry out two-way accuse, financial business risk control intensity is low.

Disclosure of Invention

The embodiment of the application aims to provide an approval risk screening method, device, equipment and storage medium thereof, which are used for solving the problems that the prior art lacks a scheme for identifying a risk application main body and a risk auditing main body simultaneously, cannot perform bidirectional control and has low risk control intensity of financial business.

In order to solve the technical problems, the embodiment of the application provides an approval risk screening method, which adopts the following technical scheme:

an approval risk screening method comprising the steps of:

acquiring history approval information of approved approval, wherein the history approval information comprises an approval chain of each piece of history approval data, approval material category information, approval content, approval request time and a request main body identifier for submitting an approval application;

identifying the approval chain and the approval material category information through a preset identification engine, and verifying the program rationality of the approval chain and the integrity of the approval material category information;

performing word recognition and semantic analysis on the approval content to obtain a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application;

according to program rationality of approval chains in the historical approval information, integrity of approval material category information, approval request time and semantic analysis results corresponding to approval contents, which are respectively corresponding to all request subjects, statistics is carried out on historical request behaviors of all request subjects and corresponding historical approval results, wherein the historical approval results comprise approval subject identifications for approval;

Performing risk screening on the historical request behaviors of all request subjects and corresponding historical approval results through a preset risk screening model, and screening out risk request subjects and risk approval subjects;

and sending the request subject identification of the risk request subject and the approval subject identification of the risk approval subject to a preset risk control end for risk control processing.

Further, the preset recognition engine includes a first recognition component and a second recognition component, and before executing the steps of recognizing the approval chain and the approval material category information by the preset recognition engine, verifying the program rationality of the approval chain and the integrity of the approval material category information, the method further includes:

according to a preset approval configuration list, determining all approval nodes in a target approval process and approval materials required by each approval node;

combining the time sequence of the target approval process, and constructing a directional comprehensive approval chain comprising the full approval nodes by taking the full approval nodes as chain nodes;

constructing a first identification component by taking the directional comprehensive approval chain as a reference approval chain;

Constructing a second identification component according to the approval materials required by each approval node;

the step of identifying the approval chain and the approval material category information through a preset identification engine, verifying the program rationality of the approval chain and the integrity of the approval material category information specifically comprises the following steps:

inputting the approval chain into the first identification component, and verifying the program rationality of the approval chain according to the standard approval chain in the first identification component;

inputting the approval material category information into the second identification component, and verifying the integrity of the approval material category according to the second identification component.

Further, the step of inputting the approval chain into the first identification component and verifying the program rationality of the approval chain according to the standard approval chain in the first identification component specifically includes:

comparing the approval chain with the standard approval chain, and identifying whether the approval chain is a sub-chain of the standard approval chain;

if the approval chain is a sub-chain of the standard approval chain, the approval chain has program rationality;

if the approval chain is not a child of the benchmark approval chain, the approval chain is not program-justified.

Further, the step of inputting the approval material category information into the second recognition component, and verifying the integrity of the approval material category according to the second recognition component specifically includes:

based on the approval chain, identifying all approval nodes corresponding to the current historical approval data as target approval nodes;

determining approval materials corresponding to each approval node in the target approval nodes through the second identification component;

determining all the approval materials which should be submitted by the current historical approval data according to the approval materials corresponding to each approval node in the target approval nodes;

comparing and identifying the type information of the approval materials corresponding to the current historical approval data obtained from the historical approval information with all the approval materials which should be submitted;

if the approval material category information corresponding to the current historical approval data obtained from the historical approval information contains all the approval materials which should be submitted, the current historical approval data has approval material category integrity;

if the approval material category information corresponding to the current historical approval data obtained from the historical approval information cannot contain all the approval materials which should be submitted, the current historical approval data does not have the approval material category integrity.

Further, the step of performing text recognition and semantic analysis on the approval content to obtain a semantic analysis result specifically includes:

identifying the approval content through an OCR character identification technology;

carrying out semantic analysis on the identified approval content according to an NLP semantic analysis technology to obtain a semantic analysis result;

after executing the steps of performing text recognition and semantic analysis on the approval content and obtaining a semantic analysis result, the method further comprises the following steps:

and determining an approval result corresponding to the current historical approval data according to the semantic analysis result.

Further, the step of counting the historical request behaviors of all the request subjects and the corresponding historical approval results according to the program rationality of the approval chain in the historical approval information, the integrity of the approval material category information, the approval request time and the semantic analysis results corresponding to the approval contents, which are respectively corresponding to all the request subjects, specifically includes:

according to the request main body identification of the submitted approval application, counting the approval request time, the number of requests with program rationality, the number of requests without program rationality, the number of requests with approval material category integrity and the number of requests without approval material category integrity of the same request main body history when submitting approval requests each time;

Counting the total times of historical submission of approval requests of the same request main body according to the number of the approval request times;

counting the number of times of approval material category integrality as the effective number of request behaviors when the same request main body historically submits approval requests according to the approval request time;

according to the approval request time, respectively corresponding approval request time when each request behavior is effective and respectively corresponding approval request time when each request behavior is ineffective are identified;

according to the request main body identification of the submitted approval application, statistics is carried out on the approval contents respectively corresponding to the same request main body history after each submission of the approval request, and according to the semantic analysis results corresponding to the approval contents, the approval results respectively corresponding to the same request main body history after each submission of the approval request are obtained;

classifying the corresponding approval results after each time of submitting an approval request according to the histories of the same request main body, and counting the passing times and the failing times of the histories according to the classification results;

according to the approval request time, respectively corresponding approval request time when each historical approval passes and respectively corresponding approval request time when each historical approval fails are counted;

According to the approval request time which corresponds to each request behavior effectively, the approval request time which corresponds to each request behavior invalidity, the approval request time which corresponds to each history approval passing, and the approval request time which corresponds to each history approval failing, the approval correct times and the approval error times of the approval behaviors are counted;

and counting the correct times of the approval behaviors, the incorrect times of the approval behaviors and the total times of the approval behaviors when the same approval main body performs approval according to the approval main body identification for approval.

Further, the step of performing risk screening on the historical request behaviors of all the request subjects and the corresponding historical approval results through a preset risk screening model to predict a risk approval request and a risk approval result specifically includes:

acquiring the total times of submitting approval requests respectively by all request main body histories and the effective times of request behaviors;

the total times of historical submission and approval requests of the current request main body and the effective times of the request behaviors corresponding to the current request main body are used as paired parameters, and the risk screening model is input;

calculating the ratio between the effective times of the request behavior and the total times of the historical submission of the approval request of the current request main body as a first ratio according to a ratio algorithm preset in the risk screening model;

If the first ratio is smaller than a preset first ratio threshold, marking the current request body as a risk request body;

obtaining the number of approval behavior errors and the total number of approval behaviors when all approval subjects respectively conduct approval;

the number of approval behavior errors and the total number of approval behaviors when the current approval main body performs approval are used as paired parameters, and the risk screening model is input;

calculating the ratio of the number of errors of the examination and approval behaviors to the total number of the examination and approval behaviors as a second ratio according to a ratio algorithm preset in the risk screening model;

and if the second ratio is larger than a preset second ratio threshold, marking the current approval body as a risk approval body.

In order to solve the technical problems, the embodiment of the application also provides an approval risk screening device, which adopts the following technical scheme:

an approval risk screening device comprising:

the historical approval information acquisition module is used for acquiring the approved historical approval information, wherein the historical approval information comprises an approval chain of each piece of historical approval data, approval material category information, approval content, approval request time and a request main body identifier for submitting an approval application;

The engine identification module is used for identifying the approval chain and the approval material category information through a preset identification engine and verifying the program rationality of the approval chain and the integrity of the approval material category information;

the approval content analysis module is used for carrying out word recognition and semantic analysis on the approval content, obtaining a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application;

the comprehensive statistics module is used for counting the historical request behaviors of all request subjects and the corresponding historical approval results according to the program rationality of an approval chain in the historical approval information, the integrity of approval material category information, the approval request time and the semantic analysis results corresponding to the approval contents, which are respectively corresponding to all request subjects, wherein the historical approval results comprise approval subject identifications for approval;

the risk screening module is used for screening the risks of the historical request behaviors of all request subjects and the corresponding historical approval results through a preset risk screening model, and screening out risk request subjects and risk approval subjects;

And the screening result processing module is used for sending the request main body identifier of the risk request main body and the approval main body identifier of the risk approval main body to a preset risk control end to perform risk control processing.

In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical schemes:

a computer device comprising a memory having stored therein computer readable instructions which when executed by the processor implement the steps of the approval risk screening method described above.

In order to solve the above technical problems, an embodiment of the present application further provides a computer readable storage medium, which adopts the following technical schemes:

a computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of an approval risk screening method as described above.

Compared with the prior art, the embodiment of the application has the following main beneficial effects:

according to the examination and approval risk screening method, history examination and approval information of the examined and approved information is obtained; identifying the approval chain and the approval material category information through a preset identification engine, and verifying the program rationality of the approval chain and the integrity of the approval material category information; performing word recognition and semantic analysis on the approval content to obtain a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application; according to program rationality of the approval chain, integrity of the class of the approved materials, approval request time and semantic analysis results corresponding to the approval contents in all the history approval information corresponding to the same request main body, statistics is carried out on history request behaviors of different request main bodies and corresponding history approval results; performing risk screening on the historical request behaviors of all request subjects and corresponding historical approval results through a preset risk screening model, and screening out risk request subjects and risk approval subjects; and sending the request subject identification of the risk request subject and the approval subject identification of the risk approval subject to a preset risk control end for risk control processing. In the loan application and credit card application business of the finance company, bidirectional control is performed by analyzing the behaviors of the request main body and the behaviors of the auditing main body, so that the illegal approval behaviors of the finance business are reduced as much as possible, and the risk control degree of the finance business is improved.

Drawings

In order to more clearly illustrate the solution of the present application, a brief description will be given below of the drawings required for the description of the embodiments of the present application, it being apparent that the drawings in the following description are some embodiments of the present application, and that other drawings may be obtained from these drawings without the exercise of inventive effort for a person of ordinary skill in the art.

FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;

FIG. 2 is a flow chart of one embodiment of an approval risk screening method according to the present application;

FIG. 3 is a flow chart of one embodiment of step 202 of FIG. 2;

FIG. 4 is a flow chart of one embodiment of step 302 shown in FIG. 3;

FIG. 5 is a flow chart of one embodiment of step 204 shown in FIG. 2;

FIG. 6 is a schematic structural view of one embodiment of an approval risk screening device according to the present application;

FIG. 7 is a schematic diagram of an embodiment of a computer device in accordance with the application.

Detailed Description

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the applications herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description of the application and the claims and the description of the drawings above are intended to cover a non-exclusive inclusion. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

In order to make the person skilled in the art better understand the solution of the present application, the technical solution of the embodiment of the present application will be clearly and completely described below with reference to the accompanying drawings.

As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the server 105 via the network 104 using the terminal devices 101, 102, 103 to receive or send messages or the like. Various communication client applications, such as a web browser application, a shopping class application, a search class application, an instant messaging tool, a mailbox client, social platform software, etc., may be installed on the terminal devices 101, 102, 103.

The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablet computers, electronic book readers, MP3 players (Moving Picture ExpertsGroup Audio Layer III, dynamic video expert compression standard audio plane 3), MP4 (Moving PictureExperts Group Audio Layer IV, dynamic video expert compression standard audio plane 4) players, laptop and desktop computers, and the like.

The server 105 may be a server providing various services, such as a background server providing support for pages displayed on the terminal devices 101, 102, 103.

It should be noted that, the approval risk screening method provided by the embodiment of the present application is generally executed by a server/terminal device, and accordingly, the approval risk screening device is generally disposed in the server/terminal device.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

With continued reference to fig. 2, a flow chart of one embodiment of an approval risk screening method according to the present application is shown. The examination and approval risk screening method comprises the following steps:

Step 201, obtaining history approval information of approved, wherein the history approval information comprises an approval chain of each piece of history approval data, approval material category information, approval content, approval request time and a request main body identifier of a submitted approval application.

In this embodiment, the approved historical approval information may be approval information of staff's daily leave in the finance company, or may be approval information of the finance company when accepting loans and credit card application services, where the approval chain is composed of approval nodes layer by layer, and the approval material category information may be different according to different service types or different approval nodes, for example, the required approval material may only need to be a leave for staff's daily leave in the finance company, and for the approval of the finance company when accepting loans and credit card application services, in general, various kinds of approval materials may relate to identity information, bank running details information, personal investigation information, warranter information, and contract text after signing, etc. The approval content comprises field information of approval passing or approval failing entered or written by an approver and approval person signature information, the approval request time generally follows the current time information of an automatic binding target request end of an approval request, and the request main body identification of the submitted approval application comprises user identification information of the submitted approval request.

Step 202, identifying the approval chain and the approval material category information through a preset identification engine, and verifying the program rationality of the approval chain and the integrity of the approval material category information.

In this embodiment, the preset recognition engine includes a first recognition component and a second recognition component, and before executing the steps of recognizing the approval chain and the approval material category information by the preset recognition engine, verifying the program rationality of the approval chain and the integrity of the approval material category information, the method further includes: according to a preset approval configuration list, determining all approval nodes in a target approval process and approval materials required by each approval node; combining the time sequence of the target approval process, and constructing a directional comprehensive approval chain comprising the full approval nodes by taking the full approval nodes as chain nodes; constructing a first identification component by taking the directional comprehensive approval chain as a reference approval chain; and constructing a second identification component according to the approval materials required by each approval node.

With continued reference to FIG. 3, FIG. 3 is a flow chart of one embodiment of step 202 shown in FIG. 2, comprising:

Step 301, inputting the approval chain into the first identification component, and verifying the program rationality of the approval chain according to a reference approval chain in the first identification component;

in this embodiment, the step of inputting the approval chain into the first identification component and verifying the program rationality of the approval chain according to the reference approval chain in the first identification component specifically includes: comparing the approval chain with the standard approval chain, and identifying whether the approval chain is a sub-chain of the standard approval chain; if the approval chain is a sub-chain of the standard approval chain, the approval chain has program rationality; if the approval chain is not a child of the benchmark approval chain, the approval chain is not program-justified.

Through constructing the basic approval chain in advance and putting the basic approval chain into the first identification component, whether the approval chain has program rationality or not can be identified conveniently through the first identification component. And the examination and approval risk investigation is convenient, and the examination and approval chain with the program rationality and the examination and approval chain without the program rationality are checked.

Step 302, inputting the approved material category information into the second recognition component, and verifying the integrity of the approved material category according to the second recognition component.

With continued reference to FIG. 4, FIG. 4 is a flow chart of one embodiment of step 302 shown in FIG. 3, including:

step 401, identifying all approval nodes corresponding to the current historical approval data based on the approval chain, and taking the approval nodes as target approval nodes;

step 402, determining, by the second identification component, approval materials corresponding to each approval node in the target approval nodes;

step 403, determining all the approval materials that should be submitted by the current historical approval data according to the approval materials corresponding to each approval node in the target approval nodes;

step 404, comparing and identifying the type information of the approved material corresponding to the current historical approval data obtained from the historical approval information with all the approved materials which should be submitted;

step 405, if the approval material category information corresponding to the current historical approval data obtained from the historical approval information includes all the approval materials that should be submitted, the current historical approval data has the approval material category integrity;

in step 406, if the approval material category information corresponding to the current historical approval data obtained from the historical approval information cannot include all the approval materials that should be submitted, the current historical approval data does not have the approval material category integrity.

By constructing a second identification component according to the approval materials required by each approval node, whether each piece of historical approval data has the integrity of the class of the approval materials is conveniently identified through the second identification component. And the method is convenient for examining and approving risk examination, and can be used for examining and approving historical examination and approval data with the integrity of the examination and approval material category and without the integrity of the examination and approval material category.

And 203, performing word recognition and semantic analysis on the approval content to acquire a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application.

In this embodiment, the step of performing text recognition and semantic analysis on the approval content to obtain a semantic analysis result specifically includes: identifying the approval content through an OCR character identification technology; and carrying out semantic analysis on the identified approval content according to an NLP semantic analysis technology to obtain a semantic analysis result.

In this embodiment, after executing the steps of performing text recognition and semantic analysis on the approval content to obtain a semantic analysis result, the method further includes: and determining an approval result corresponding to the current historical approval data according to the semantic analysis result.

Specifically, the approval content is identified through the OCR word recognition technology, that is, approval signature information of an approval subject is identified through the OCR word recognition technology, including approval person signature information or seal information, and specific approval content, for example, approval passing or approval failing, is identified. And carrying out semantic analysis on the identified approval content according to an NLP semantic analysis technology, namely analyzing whether the approval passes or not.

Step 204, according to the program rationality of the approval chain, the integrity of the approval material category information, the approval request time and the semantic analysis result corresponding to the approval content in the history approval information respectively corresponding to all the request subjects, the history request behaviors of all the request subjects and the corresponding history approval results are counted, wherein the history approval results comprise approval subject identifications for approval.

With continued reference to fig. 5, fig. 5 is a flow chart of one embodiment of step 204 shown in fig. 2, comprising:

step 501, counting the approval request time, the number of requests with program rationality, the number of requests without program rationality, the number of requests with approval material type integrity and the number of requests without approval material type integrity when the same request main body history submits approval requests each time according to the request main body identifier of the submitted approval application;

Step 502, counting the total times of historical submission of approval requests of the same request main body according to the number of approval request time;

step 503, counting the number of times of checking the integrity of the material class as the effective number of request actions, wherein the number of times of checking the integrity of the material class is counted as the effective number of times of request actions when the same request main body historically submits the checking request according to the checking request time;

step 504, according to the approval request time, identifying the approval request time corresponding to each valid request behavior and the approval request time corresponding to each invalid request behavior;

step 505, according to the request subject identification of the submitted approval application, statistics is carried out on the approval contents respectively corresponding to the same request subject history after each submission of the approval request, and according to the semantic analysis results corresponding to the approval contents, the approval results respectively corresponding to the same request subject history after each submission of the approval request are obtained;

step 506, classifying according to the corresponding approval results after each time of submitting the approval request of the same request main body history, and counting the passing times and the failing times of the historical approval according to the classification results;

step 507, counting the approval request time corresponding to each passing of the historical approval according to the approval request time, and counting the approval request time corresponding to each failing of the historical approval;

Step 508, counting the correct number of the approval behaviors and the incorrect number of the approval behaviors according to the approval request time corresponding to each request behavior effective, the approval request time corresponding to each invalid request behavior, the approval request time corresponding to each history approval passing, and the approval request time corresponding to each history approval failing;

step 509, counting the correct number of approval behaviors, the incorrect number of approval behaviors and the total number of approval behaviors when the same approval subject performs approval according to the approval subject identification for approval.

By adopting a comprehensive analysis and statistics method, the total times of historical submission of the approval request and the effective times of the request behaviors of the same request main body are counted, and the correct times, the incorrect times and the total times of the approval behaviors are convenient for a subsequent risk screening model to carry out risk screening according to the counted times when different approval main bodies carry out approval. In particular, in the loan application and credit card application business of a finance company, double-layer control is performed by analyzing the behaviors of a request main body and the behaviors of an auditing main body, the probability of fraudulent credit and illegal credit card application is reduced as much as possible, and the risk control degree of the finance business is improved.

Step 205, performing risk screening on the historical request behaviors and the corresponding historical approval results of all the request subjects through a preset risk screening model, and predicting a risk approval request and a risk approval result.

In this embodiment, the step of performing risk screening on the historical request behaviors of all the request subjects and the corresponding historical approval results through the preset risk screening model to predict a risk approval request and a risk approval result specifically includes: acquiring the total times of submitting approval requests respectively by all request main body histories and the effective times of request behaviors; the total times of historical submission and approval requests of the current request main body and the effective times of the request behaviors corresponding to the current request main body are used as paired parameters, and the risk screening model is input; calculating the ratio between the effective times of the request behavior and the total times of the historical submission of the approval request of the current request main body as a first ratio according to a ratio algorithm preset in the risk screening model; if the first ratio is smaller than a preset first ratio threshold, marking the current request body as a risk request body; if the first ratio is not smaller than a preset first ratio threshold, marking the current request body as a non-risk request body; obtaining the number of approval behavior errors and the total number of approval behaviors when all approval subjects respectively conduct approval; the number of approval behavior errors and the total number of approval behaviors when the current approval main body performs approval are used as paired parameters, and the risk screening model is input; calculating the ratio of the number of errors of the examination and approval behaviors to the total number of the examination and approval behaviors as a second ratio according to a ratio algorithm preset in the risk screening model; if the second ratio is larger than a preset second ratio threshold, marking the current approval main body as a risk approval main body; and if the second ratio is not greater than a preset second ratio threshold, marking the current approval body as a non-risk approval body.

By adopting a ratio algorithm and combining the total times of historical submission of approval requests and the effective times of request behaviors of the same request main body, the number of incorrect approval behaviors and the total number of approval behaviors of different approval main bodies when the different approval main bodies perform approval are convenient for a subsequent risk screening model to perform risk screening according to the statistical times. In particular, in the loan application and credit card application business of a finance company, bidirectional control is performed by analyzing the behaviors of a request main body and the behaviors of an auditing main body, so that the probability of fraudulent credit and illegal credit card application is reduced as much as possible, and the risk control degree of the finance business is improved.

Step 206, the request subject identifier of the risk request subject and the approval subject identifier of the risk approval subject are sent to a preset risk control end to perform risk control processing.

The method comprises the steps of obtaining historical approval information of approved approval; identifying the approval chain and the approval material category information through a preset identification engine, and verifying the program rationality of the approval chain and the integrity of the approval material category information; performing word recognition and semantic analysis on the approval content to obtain a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application; according to program rationality of the approval chain, integrity of the class of the approved materials, approval request time and semantic analysis results corresponding to the approval contents in all the history approval information corresponding to the same request main body, statistics is carried out on history request behaviors of different request main bodies and corresponding history approval results; performing risk screening on the historical request behaviors of all request subjects and corresponding historical approval results through a preset risk screening model, and screening out risk request subjects and risk approval subjects; and sending the request subject identification of the risk request subject and the approval subject identification of the risk approval subject to a preset risk control end for risk control processing. In the loan application and credit card application business of the finance company, bidirectional control is performed by analyzing the behaviors of the request main body and the behaviors of the auditing main body, so that the illegal approval behaviors of the finance business are reduced as much as possible, and the risk control degree of the finance business is improved.

The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.

Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

In the embodiment of the application, the history approval information of the approved information is obtained; screening all the historical approval information corresponding to the same request main body, and carrying out comprehensive analysis; performing risk screening on the historical request behaviors of different request subjects and the corresponding historical approval results, and screening out risk request subjects and risk approval subjects; and sending the request subject identification of the risk request subject and the approval subject identification of the risk approval subject to a preset risk control end for risk control processing. In the loan application and credit card application business of the finance company, the behaviors of the request main body and the behaviors of the auditing main body are analyzed to carry out bidirectional control, so that the illegal approval behaviors of the finance business are reduced as much as possible, the risk control degree of the finance business is improved, the history approval information which is finished by the examination and approval can be comprehensively analyzed and calculated in an artificial intelligence mode, and the manual calculation amount is reduced.

With further reference to fig. 6, as an implementation of the method shown in fig. 2, the present application provides an embodiment of an approval risk screening apparatus, which corresponds to the method embodiment shown in fig. 2, and which is particularly applicable to various electronic devices.

As shown in fig. 6, the approval risk screening apparatus 600 according to the present embodiment includes: a historical approval information acquisition module 601, an engine identification module 602, an approval content analysis module 603, a comprehensive statistics module 604, a risk screening module 605 and a screening result processing module 606. Wherein:

the historical approval information obtaining module 601 is configured to obtain approved historical approval information, where the historical approval information includes an approval chain of each piece of historical approval data, approval material category information, approval content, approval request time, and a request body identifier of a submitted approval application;

the engine identification module 602 is configured to identify the approval chain and the approval material class information by using a preset identification engine, and verify the program rationality of the approval chain and the integrity of the approval material class information;

the approval content analysis module 603 is configured to perform text recognition and semantic analysis on the approval content, obtain a semantic analysis result, and screen out historical approval information corresponding to the same request main body according to a request main body identifier of the submitted approval application;

The comprehensive statistics module 604 is configured to count historical request behaviors of all request principals and corresponding historical approval results according to program rationality of approval chains in the historical approval information, integrity of approval material class information, approval request time and semantic analysis results corresponding to approval contents, where the historical approval results include approval principal identifiers for approval;

the risk screening module 605 is configured to screen, through a preset risk screening model, the historical request behaviors of all the request subjects and the corresponding historical approval results, and screen out risk request subjects and risk approval subjects;

the screening result processing module 606 is configured to send the request body identifier of the risk request body and the approval body identifier of the risk approval body to a preset risk control end, so as to perform risk control processing.

The method comprises the steps of obtaining historical approval information of approved approval; identifying the approval chain and the approval material category information through a preset identification engine, and verifying the program rationality of the approval chain and the integrity of the approval material category information; performing word recognition and semantic analysis on the approval content to obtain a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application; according to program rationality of the approval chain, integrity of the class of the approved materials, approval request time and semantic analysis results corresponding to the approval contents in all the history approval information corresponding to the same request main body, statistics is carried out on history request behaviors of different request main bodies and corresponding history approval results; performing risk screening on the historical request behaviors of all request subjects and corresponding historical approval results through a preset risk screening model, and screening out risk request subjects and risk approval subjects; and sending the request subject identification of the risk request subject and the approval subject identification of the risk approval subject to a preset risk control end for risk control processing. In the loan application and credit card application business of the finance company, bidirectional control is performed by analyzing the behaviors of the request main body and the behaviors of the auditing main body, so that the illegal approval behaviors of the finance business are reduced as much as possible, and the risk control degree of the finance business is improved.

Those skilled in the art will appreciate that implementing all or part of the above described embodiment methods may be accomplished by computer readable instructions, stored on a computer readable storage medium, that the program when executed may comprise the steps of embodiments of the methods described above. The storage medium may be a nonvolatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.

In order to solve the technical problems, the embodiment of the application also provides computer equipment. Referring specifically to fig. 7, fig. 7 is a basic structural block diagram of a computer device according to the present embodiment.

The computer device 7 comprises a memory 7a, a processor 7b, a network interface 7c communicatively connected to each other via a system bus. It should be noted that only a computer device 7 having components 7a-7c is shown in the figures, but it should be understood that not all of the illustrated components need be implemented, and that more or fewer components may alternatively be implemented. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculations and/or information processing in accordance with predetermined or stored instructions, the hardware of which includes, but is not limited to, microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASICs), programmable gate arrays (fields-Programmable Gate Array, FPGAs), digital processors (Digital Signal Processor, DSPs), embedded devices, etc.

The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer equipment can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.

The memory 7a includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the storage 7a may be an internal storage unit of the computer device 7, such as a hard disk or a memory of the computer device 7. In other embodiments, the memory 7a may also be an external storage device of the computer device 7, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 7. Of course, the memory 7a may also comprise both an internal memory unit of the computer device 7 and an external memory device. In this embodiment, the memory 7a is typically used to store an operating system and various application software installed on the computer device 7, such as computer readable instructions of an approval risk screening method. Further, the memory 7a may be used to temporarily store various types of data that have been output or are to be output.

The processor 7b may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 7b is typically used to control the overall operation of the computer device 7. In this embodiment, the processor 7b is configured to execute computer readable instructions stored in the memory 7a or process data, such as computer readable instructions for executing the approval risk screening method.

The network interface 7c may comprise a wireless network interface or a wired network interface, which network interface 7c is typically used for establishing a communication connection between the computer device 7 and other electronic devices.

The computer equipment provided by the embodiment belongs to the technical field of financial science and technology, and is applied to a financial business approval scene. The method comprises the steps of obtaining historical approval information of approved approval; identifying the approval chain and the approval material category information through a preset identification engine, and verifying the program rationality of the approval chain and the integrity of the approval material category information; performing word recognition and semantic analysis on the approval content to obtain a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application; according to program rationality of the approval chain, integrity of the class of the approved materials, approval request time and semantic analysis results corresponding to the approval contents in all the history approval information corresponding to the same request main body, statistics is carried out on history request behaviors of different request main bodies and corresponding history approval results; performing risk screening on the historical request behaviors of all request subjects and corresponding historical approval results through a preset risk screening model, and screening out risk request subjects and risk approval subjects; and sending the request subject identification of the risk request subject and the approval subject identification of the risk approval subject to a preset risk control end for risk control processing. In the loan application and credit card application business of the finance company, bidirectional control is performed by analyzing the behaviors of the request main body and the behaviors of the auditing main body, so that the illegal approval behaviors of the finance business are reduced as much as possible, and the risk control degree of the finance business is improved.

The present application also provides another embodiment, namely, a computer readable storage medium storing computer readable instructions executable by a processor to cause the processor to perform the steps of the method for screening for approval risk as described above.

The computer readable storage medium provided by the embodiment belongs to the technical field of financial science and technology, and is applied to a financial business approval scene. The method comprises the steps of obtaining historical approval information of approved approval; identifying the approval chain and the approval material category information through a preset identification engine, and verifying the program rationality of the approval chain and the integrity of the approval material category information; performing word recognition and semantic analysis on the approval content to obtain a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application; according to program rationality of the approval chain, integrity of the class of the approved materials, approval request time and semantic analysis results corresponding to the approval contents in all the history approval information corresponding to the same request main body, statistics is carried out on history request behaviors of different request main bodies and corresponding history approval results; performing risk screening on the historical request behaviors of all request subjects and corresponding historical approval results through a preset risk screening model, and screening out risk request subjects and risk approval subjects; and sending the request subject identification of the risk request subject and the approval subject identification of the risk approval subject to a preset risk control end for risk control processing. In the loan application and credit card application business of the finance company, bidirectional control is performed by analyzing the behaviors of the request main body and the behaviors of the auditing main body, so that the illegal approval behaviors of the finance business are reduced as much as possible, and the risk control degree of the finance business is improved.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present application.

It is apparent that the above-described embodiments are only some embodiments of the present application, but not all embodiments, and the preferred embodiments of the present application are shown in the drawings, which do not limit the scope of the patent claims. This application may be embodied in many different forms, but rather, embodiments are provided in order to provide a thorough and complete understanding of the present disclosure. Although the application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing description, or equivalents may be substituted for elements thereof. All equivalent structures made by the content of the specification and the drawings of the application are directly or indirectly applied to other related technical fields, and are also within the scope of the application.

Claims (10)

1. An approval risk screening method, comprising the steps of:

acquiring history approval information of approved approval, wherein the history approval information comprises an approval chain of each piece of history approval data, approval material category information, approval content, approval request time and a request main body identifier for submitting an approval application;

identifying the approval chain and the approval material category information through a preset identification engine, and verifying the program rationality of the approval chain and the integrity of the approval material category information;

performing word recognition and semantic analysis on the approval content to obtain a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application;

according to program rationality of approval chains in the historical approval information, integrity of approval material category information, approval request time and semantic analysis results corresponding to approval contents, which are respectively corresponding to all request subjects, statistics is carried out on historical request behaviors of all request subjects and corresponding historical approval results, wherein the historical approval results comprise approval subject identifications for approval;

Performing risk screening on the historical request behaviors of all request subjects and corresponding historical approval results through a preset risk screening model, and screening out risk request subjects and risk approval subjects;

and sending the request subject identification of the risk request subject and the approval subject identification of the risk approval subject to a preset risk control end for risk control processing.

2. The method of claim 1, wherein the predetermined recognition engine includes a first recognition component and a second recognition component, and wherein prior to performing the steps of recognizing the approval chain, the approval material category information, verifying the procedural rationality of the approval chain, and the integrity of the approval material category information by the predetermined recognition engine, the method further comprises:

according to a preset approval configuration list, determining all approval nodes in a target approval process and approval materials required by each approval node;

combining the time sequence of the target approval process, and constructing a directional comprehensive approval chain comprising the full approval nodes by taking the full approval nodes as chain nodes;

constructing a first identification component by taking the directional comprehensive approval chain as a reference approval chain;

Constructing a second identification component according to the approval materials required by each approval node;

the step of identifying the approval chain and the approval material category information through a preset identification engine, verifying the program rationality of the approval chain and the integrity of the approval material category information specifically comprises the following steps:

inputting the approval chain into the first identification component, and verifying the program rationality of the approval chain according to the standard approval chain in the first identification component;

inputting the approval material category information into the second identification component, and verifying the integrity of the approval material category according to the second identification component.

3. The method for screening approval risk according to claim 2, wherein the step of inputting the approval chain into the first recognition component and verifying the program rationality of the approval chain according to the standard approval chain in the first recognition component specifically comprises:

comparing the approval chain with the standard approval chain, and identifying whether the approval chain is a sub-chain of the standard approval chain;

if the approval chain is a sub-chain of the standard approval chain, the approval chain has program rationality;

If the approval chain is not a child of the benchmark approval chain, the approval chain is not program-justified.

4. The method for screening risk of approval according to claim 2, wherein said step of inputting said approval material class information into said second identification component and verifying the integrity of said approval material class according to said second identification component comprises:

based on the approval chain, identifying all approval nodes corresponding to the current historical approval data as target approval nodes;

determining approval materials corresponding to each approval node in the target approval nodes through the second identification component;

determining all the approval materials which should be submitted by the current historical approval data according to the approval materials corresponding to each approval node in the target approval nodes;

comparing and identifying the type information of the approval materials corresponding to the current historical approval data obtained from the historical approval information with all the approval materials which should be submitted;

if the approval material category information corresponding to the current historical approval data obtained from the historical approval information contains all the approval materials which should be submitted, the current historical approval data has approval material category integrity;

If the approval material category information corresponding to the current historical approval data obtained from the historical approval information cannot contain all the approval materials which should be submitted, the current historical approval data does not have the approval material category integrity.

5. The method for screening approval risk according to claim 1, wherein the step of performing text recognition and semantic analysis on the approval content to obtain a semantic analysis result specifically comprises:

identifying the approval content through an OCR character identification technology;

carrying out semantic analysis on the identified approval content according to an NLP semantic analysis technology to obtain a semantic analysis result;

after executing the steps of performing text recognition and semantic analysis on the approval content and obtaining a semantic analysis result, the method further comprises the following steps:

and determining an approval result corresponding to the current historical approval data according to the semantic analysis result.

6. The method for screening approval risk according to claim 1, wherein the step of counting the historical request behaviors of all the request subjects and the corresponding historical approval results according to the program rationality of the approval chain, the integrity of the approval material category information, the approval request time and the semantic analysis results corresponding to the approval contents in the historical approval information corresponding to all the request subjects respectively specifically comprises:

According to the request main body identification of the submitted approval application, counting the approval request time, the number of requests with program rationality, the number of requests without program rationality, the number of requests with approval material category integrity and the number of requests without approval material category integrity of the same request main body history when submitting approval requests each time;

counting the total times of historical submission of approval requests of the same request main body according to the number of the approval request times;

counting the number of times of approval material category integrality as the effective number of request behaviors when the same request main body historically submits approval requests according to the approval request time;

according to the approval request time, respectively corresponding approval request time when each request behavior is effective and respectively corresponding approval request time when each request behavior is ineffective are identified;

according to the request main body identification of the submitted approval application, statistics is carried out on the approval contents respectively corresponding to the same request main body history after each submission of the approval request, and according to the semantic analysis results corresponding to the approval contents, the approval results respectively corresponding to the same request main body history after each submission of the approval request are obtained;

Classifying the corresponding approval results after each time of submitting an approval request according to the histories of the same request main body, and counting the passing times and the failing times of the histories according to the classification results;

according to the approval request time, respectively corresponding approval request time when each historical approval passes and respectively corresponding approval request time when each historical approval fails are counted;

according to the approval request time which corresponds to each request behavior effectively, the approval request time which corresponds to each request behavior invalidity, the approval request time which corresponds to each history approval passing, and the approval request time which corresponds to each history approval failing, the approval correct times and the approval error times of the approval behaviors are counted;

and counting the correct times of the approval behaviors, the incorrect times of the approval behaviors and the total times of the approval behaviors when the same approval main body performs approval according to the approval main body identification for approval.

7. The method for screening risk of approval according to claim 6, wherein the step of performing risk screening on the historical request behaviors of all the request subjects and the corresponding historical approval results through a preset risk screening model to predict a risk approval request and a risk approval result specifically comprises the steps of:

Acquiring the total times of submitting approval requests respectively by all request main body histories and the effective times of request behaviors;

the total times of historical submission and approval requests of the current request main body and the effective times of the request behaviors corresponding to the current request main body are used as paired parameters, and the risk screening model is input;

calculating the ratio between the effective times of the request behavior and the total times of the historical submission of the approval request of the current request main body as a first ratio according to a ratio algorithm preset in the risk screening model;

if the first ratio is smaller than a preset first ratio threshold, marking the current request body as a risk request body;

obtaining the number of approval behavior errors and the total number of approval behaviors when all approval subjects respectively conduct approval;

the number of approval behavior errors and the total number of approval behaviors when the current approval main body performs approval are used as paired parameters, and the risk screening model is input;

calculating the ratio of the number of errors of the examination and approval behaviors to the total number of the examination and approval behaviors as a second ratio according to a ratio algorithm preset in the risk screening model;

and if the second ratio is larger than a preset second ratio threshold, marking the current approval body as a risk approval body.

8. An approval risk screening device, comprising:

the historical approval information acquisition module is used for acquiring the approved historical approval information, wherein the historical approval information comprises an approval chain of each piece of historical approval data, approval material category information, approval content, approval request time and a request main body identifier for submitting an approval application;

the engine identification module is used for identifying the approval chain and the approval material category information through a preset identification engine and verifying the program rationality of the approval chain and the integrity of the approval material category information;

the approval content analysis module is used for carrying out word recognition and semantic analysis on the approval content, obtaining a semantic analysis result, and screening out historical approval information corresponding to the same request main body according to the request main body identification of the submitted approval application;

the comprehensive statistics module is used for counting the historical request behaviors of all request subjects and the corresponding historical approval results according to the program rationality of an approval chain in the historical approval information, the integrity of approval material category information, the approval request time and the semantic analysis results corresponding to the approval contents, which are respectively corresponding to all request subjects, wherein the historical approval results comprise approval subject identifications for approval;

The risk screening module is used for screening the risks of the historical request behaviors of all request subjects and the corresponding historical approval results through a preset risk screening model, and screening out risk request subjects and risk approval subjects;

and the screening result processing module is used for sending the request main body identifier of the risk request main body and the approval main body identifier of the risk approval main body to a preset risk control end to perform risk control processing.

9. A computer device comprising a memory having stored therein computer readable instructions which when executed by the processor implement the steps of the method of screening for approval risk of any one of claims 1 to 7.

10. A computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the approval risk screening method of any of claims 1 to 7.