CN112819609A - Risk assessment method, apparatus, computer-readable storage medium, and program product - Google Patents

Risk assessment method, apparatus, computer-readable storage medium, and program product Download PDF

Info

Publication number
CN112819609A
CN112819609A CN202110209854.0A CN202110209854A CN112819609A CN 112819609 A CN112819609 A CN 112819609A CN 202110209854 A CN202110209854 A CN 202110209854A CN 112819609 A CN112819609 A CN 112819609A
Authority
CN
China
Prior art keywords
analysis data
risk
evaluation object
information
determining
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110209854.0A
Other languages
Chinese (zh)
Inventor
陈俊霖
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
WeBank Co Ltd
Original Assignee
WeBank Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by WeBank Co Ltd filed Critical WeBank Co Ltd
Priority to CN202110209854.0A priority Critical patent/CN112819609A/en
Publication of CN112819609A publication Critical patent/CN112819609A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/03—Credit; Loans; Processing thereof
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31—User authentication
    • G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints

Abstract

The application provides a risk assessment method, a device, a computer readable storage medium and a program product, wherein the method comprises the following steps: responding to the acquired first request for risk assessment, and outputting first interaction information for determining character analysis data; acquiring a first interaction result determined by an evaluation object based on the first interaction information, and acquiring first expression information of the evaluation object when the first interaction result is determined; determining character analysis data of the evaluation object based on the first interaction result and the first expression information; acquiring risk analysis data, and determining a risk assessment result based on the risk analysis data and the personality analysis data. By analyzing the characters of the evaluation object and judging the character tendency of the evaluation object from the psychological aspect, the accuracy of the risk evaluation result can be improved, the risk of financial credit business can be reduced, and the bad account loss can be reduced.

Description

Risk assessment method, apparatus, computer-readable storage medium, and program product
Technical Field
The present application relates to the field of financial economy technology, and relates to, but is not limited to, a risk assessment method, apparatus, computer-readable storage medium, and program product.
Background
In recent years, with the rapid development of financial credit business, the financial industry has been greatly lost due to overdue and default while the masses are popularized. The strength of the risk management capability can reduce the occurrence of overdue and default conditions to a great extent.
The current financial credit risk assessment method mainly relies on experience of business personnel to carry out face-to-face assessment, so that the assessment efficiency is low and the labor cost is high; or the judgment is carried out based on various set business rules in a risk assessment Application program (App, Application), and only the risk assessment can be carried out on the objective information of the loan user, so that the risk assessment result is inaccurate, and the financial credit business risk is higher.
Disclosure of Invention
Embodiments of the present application provide a risk assessment method, apparatus, device, computer-readable storage medium, and computer program product, which can improve accuracy of a risk assessment result and reduce a risk of a financial credit transaction.
The technical scheme of the embodiment of the application is realized as follows:
the embodiment of the application provides a risk assessment method, which comprises the following steps:
responding to the acquired first request for risk assessment, and outputting first interaction information for determining character analysis data;
acquiring a first interaction result determined by an evaluation object based on the first interaction information, and acquiring first expression information of the evaluation object when the first interaction result is determined;
determining character analysis data of the evaluation object based on the first interaction result and the first expression information;
acquiring risk analysis data, and determining a risk assessment result based on the risk analysis data and the personality analysis data.
The embodiment of the application provides a risk assessment device, the device includes:
the output module is used for responding to the acquired first request for risk assessment and outputting first interaction information for determining character analysis data;
the first acquisition module is used for acquiring a first interaction result determined by the evaluation object based on the first interaction information;
the second acquisition module is used for acquiring first expression information of the evaluation object when the first interaction result is determined;
a first determination module, configured to determine personality analysis data of the evaluation object based on the first interaction result and the first expression information;
the third acquisition module is used for acquiring risk analysis data;
a second determination module to determine a risk assessment result based on the risk analysis data and the personality analysis data.
An embodiment of the present application provides a risk assessment device, including:
a memory for storing executable instructions;
and the processor is used for realizing the method provided by the embodiment of the application when executing the executable instructions stored in the memory.
Embodiments of the present application provide a computer-readable storage medium, where executable instructions are stored on the computer-readable storage medium, and when the computer-readable storage medium is executed by a processor, the computer-readable storage medium implements a method provided by embodiments of the present application.
Embodiments of the present application provide a computer program product, which includes a computer program, and when the computer program is executed by a processor, the computer program implements the method provided by the embodiments of the present application.
The embodiment of the application has the following beneficial effects:
in the risk assessment method provided by the embodiment of the application, a first request for risk assessment is obtained first; outputting first interaction information for determining character analysis data in response to the first request; then acquiring a first interaction result determined by the evaluation object based on the first interaction information, and acquiring first expression information of the evaluation object when the first interaction result is determined; determining character analysis data of the evaluation object based on the first interaction result and the first expression information; and then acquiring risk analysis data, and finally determining a risk evaluation result based on the risk analysis data and the personality analysis data, so that a personality test is added in the risk evaluation process, and the personality tendency of an evaluation object can be judged from a psychological angle, so that the accuracy of the risk evaluation result can be improved, the risk of financial credit business can be reduced, and the bad account loss can be reduced.
Drawings
Fig. 1A is a schematic network architecture diagram of a risk assessment method according to an embodiment of the present disclosure;
fig. 1B is a schematic diagram of another network architecture of the risk assessment method according to the embodiment of the present application;
FIG. 2 is a schematic diagram of a structure of a risk assessment device provided in an embodiment of the present application;
fig. 3 is a schematic flow chart of an implementation of a risk assessment method according to an embodiment of the present application;
fig. 4 is a schematic flow chart of another implementation of the risk assessment method according to the embodiment of the present application;
fig. 5 is a schematic flowchart of another implementation of the risk assessment method according to the embodiment of the present application;
fig. 6 is a schematic flow chart of another implementation of the risk assessment method according to the embodiment of the present application;
fig. 7 is a schematic flow chart of another implementation of the risk assessment method according to the embodiment of the present application.
Detailed Description
In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be described in further detail with reference to the attached drawings, the described embodiments should not be considered as limiting the present application, and all other embodiments obtained by a person of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.
In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict.
In the following description, references to the terms "first \ second \ third" are only used to distinguish similar objects and do not denote a particular order, but rather the terms "first \ second \ third" are used to interchange specific orders or sequences, where permissible, so that the embodiments of the present application described herein can be practiced in other than the order shown or described herein.
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 herein is for the purpose of describing embodiments of the present application only and is not intended to be limiting of the application.
Before further detailed description of the embodiments of the present application, terms and expressions referred to in the embodiments of the present application will be described, and the terms and expressions referred to in the embodiments of the present application will be used for the following explanation.
1) The multi-mode, i.e. multi-mode biological recognition, is to integrate or fuse two or more biological recognition technologies, and to make the authentication and recognition process more accurate and safe by using the unique advantages of the multiple biological recognition technologies and combining the data fusion technology. The main difference from the traditional single biological identification mode is that the multi-mode biological identification technology can acquire different biological characteristics (such as fingerprints, finger veins, human faces, iris images and the like) through an independent or integrated acquisition device with multiple acquisition modes, and perform identification and authentication by analyzing and judging characteristic values of the multiple biological identification modes.
2) The digital human is a product of the integration of information science and life science, carries out virtual simulation on the shapes and functions of the human body at different levels by using an information science method, comprises four cross and overlapped development stages, a visible human, a physical human, a physiological human and an intelligent human, finally establishes a multidisciplinary and multilevel digital model and achieves the accurate simulation of the human body from the micro to the macro.
3) The multi-modal digital person refers to a digital person with the ability of real people to watch, listen, speak, write and know and logic, is not only a one-way interactive digital person, but also a two-way multi-modal interactive digital person, collects information input by a user through a camera, a microphone and other collection devices, and the digital person can identify the identity of the person and understand the language of the person. The emotion details of people are identified through sound, image and semantic technologies, the accuracy of semantic understanding is improved through the identification of gesture instructions, and corresponding responses are made.
4) Risk Assessment (Risk Assessment) refers to the task of quantitatively assessing the possibility of impact and loss on various aspects of people's life, property, etc. before or after (but not before) a Risk event occurs. That is, risk assessment quantifies the likely degree of impact or loss associated with an event or thing.
5) Wind management, or risk control, refers to the risk manager taking various measures and methods to eliminate or reduce the various possibilities of occurrence of a risk event, or the risk controller reducing the losses incurred when a risk event occurs. The four basic methods of risk control are: risk avoidance, loss control, risk transfer, and risk preservation.
An exemplary application of the apparatus implementing the embodiment of the present application is described below, and the apparatus provided in the embodiment of the present application may be implemented as a terminal device. In the following, exemplary applications covering terminal devices when the apparatus is implemented as a terminal device will be explained.
Fig. 1A is a schematic diagram of a network architecture of a risk assessment method according to an embodiment of the present disclosure, and as shown in fig. 1A, the network architecture at least includes a risk assessment device 100. To support an exemplary application, the risk assessment apparatus 100 is an apparatus capable of performing risk assessment and supporting human-computer interaction, and may be a server, or may be an apparatus such as a desktop computer, a laptop computer, a mobile phone (mobile phone), or a tablet computer.
When the risk assessment is carried out on an assessment object, a first request for carrying out the risk assessment is obtained based on an operation instruction triggered by a user, wherein when the user assesses the own risk, the user is the assessment object; when a user evaluates the risk of another person, the user is a different person from the evaluation object. After acquiring the first request, the risk assessment device 100 responds to the first request, and outputs first interaction information through voice and/or characters, wherein the first interaction information is used for determining character analysis data of an assessment object. The user determines a first interaction result based on the output first interaction information and inputs the first interaction result to the risk assessment apparatus 100. The risk assessment apparatus 100 acquires the first interaction result and acquires first expression information when the assessment target determines the first interaction result. Then, the risk assessment apparatus 100 determines the personality analysis data of the assessment target using the first interaction result and the first expression information. And after the risk analysis data is obtained, combining the character analysis data and the risk analysis data to obtain a risk evaluation result. According to the embodiment of the application, the personality analysis is carried out on the evaluation object, the personality tendency of the evaluation object is judged from a psychological angle, the accuracy of a risk evaluation result can be improved, the risk of financial credit business is reduced, and then bad account loss is reduced.
Fig. 1B is a schematic diagram of another network architecture of the risk assessment method according to the embodiment of the present application, and as shown in fig. 1B, the network architecture at least includes a risk assessment device 100, a terminal 200, and a network 300. To support an exemplary application, the risk assessment device 100 is a device capable of performing risk assessment, and may be a server, or may be a desktop computer, a notebook computer, an intelligent terminal, or the like. The terminal 200 is a terminal held by the evaluation object, and may be a mobile phone (mobile phone), a tablet computer, a notebook computer, or the like. The risk assessment device 100 is connected to the terminal 200 via a network 300, and the network 300 may be a wide area network or a local area network, or a combination of both, and uses wireless or wired links to achieve data transmission.
When an evaluation target (i.e., a user who holds the terminal 200) performs a risk evaluation in advance, the terminal 200 transmits a first request for performing a risk evaluation to the risk evaluation apparatus 100. The risk assessment apparatus 100 receives and responds to the first request, and sends first interaction information to the terminal 200, where the first interaction information is used for determining the character analysis data. After receiving the first interactive information, the terminal 200 outputs the first interactive information through voice and/or text, and the evaluation object determines a first interactive result through voice and/or text. The terminal 200 sends the first interaction result to the risk assessment apparatus 100, and simultaneously obtains first expression information of the assessment object when determining the first interaction result, and sends the first expression information to the risk assessment apparatus 100. The risk assessment device 100 determines the personality analysis data of the assessment object according to the first interaction result and the first expression information, and obtains a risk assessment result by combining the personality analysis data and the risk analysis data after obtaining the risk analysis data. According to the embodiment of the application, the personality analysis is carried out on the evaluation object, the personality tendency of the evaluation object is judged from a psychological angle, the accuracy of a risk evaluation result can be improved, the risk of financial credit business is reduced, and then bad account loss is reduced.
The apparatus provided in the embodiments of the present application may be implemented as hardware or a combination of hardware and software, and various exemplary implementations of the apparatus provided in the embodiments of the present application are described below.
Other exemplary configurations of the risk assessment device 100 are envisioned from the exemplary configuration of the risk assessment device 100 shown in FIG. 2, and thus the configurations described herein should not be considered limiting, e.g., some of the components described below may be omitted, or components not described below may be added to accommodate the particular needs of certain applications.
The risk assessment apparatus 100 shown in fig. 2 includes: at least one processor 110, memory 140, at least one network interface 120, and a user interface 130. Each of the components in the risk assessment device 100 are coupled together by a bus system 150. It will be appreciated that the bus system 150 is used to enable communications among the components of the connection. The bus system 150 includes a power bus, a control bus, and a status signal bus in addition to a data bus. For clarity of illustration, however, the various buses are labeled as bus system 150 in fig. 2.
The user interface 130 may include a display, a keyboard, a mouse, a touch-sensitive pad, a touch screen, and the like.
The memory 140 may be either volatile memory or nonvolatile memory, and may also include both volatile and nonvolatile memory. The nonvolatile Memory may be a Read Only Memory (ROM). The volatile Memory may be a Random Access Memory (RAM). The memory 140 described in embodiments herein is intended to comprise any suitable type of memory.
The memory 140 in embodiments of the present application is capable of storing data to support the operation of the risk assessment device 100. Examples of such data include: any computer program for operating on risk assessment device 100, such as an operating system and application programs. The operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, and the like, and is used for implementing various basic services and processing hardware-based tasks. The application program may include various application programs.
As an example of the method provided by the embodiment of the present application implemented by software, the method provided by the embodiment of the present application may be directly embodied as a combination of software modules executed by the processor 110, the software modules may be located in a storage medium located in the memory 140, and the processor 110 reads executable instructions included in the software modules in the memory 140, and completes the method provided by the embodiment of the present application in combination with necessary hardware (for example, including the processor 110 and other components connected to the bus 150).
By way of example, the Processor 110 may be an integrated circuit chip having Signal processing capabilities, such as a general purpose Processor, a Digital Signal Processor (DSP), or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or the like, wherein the general purpose Processor may be a microprocessor or any conventional Processor or the like.
The risk assessment is a process of analyzing and determining risks before a risk event is finished, in the field of financial credit business, risk assessment results are obtained by performing risk assessment on objective information of loan users in related technologies, and the risk assessment results include that some loan users objectively have repayment capability, but do not actively repay subjectively and have malicious debt, so that the financial credit business has higher risk. In order to solve the problem, the risk assessment method provided by the embodiment of the present application will be described in conjunction with exemplary applications and implementations of the terminal provided by the embodiment of the present application.
Fig. 3 is a schematic implementation flow diagram of a risk assessment method provided in an embodiment of the present application, which is applied to a risk assessment device of a network architecture shown in fig. 1A, and will be described with reference to the steps shown in fig. 3.
Step S301, responding to the acquired first request for risk assessment, and outputting first interaction information for determining character analysis data.
When risk assessment is performed on an assessment object with loan requirements or investment financing requirements, the risk assessment device obtains a first request for performing the risk assessment, where the risk assessment device may obtain the first request based on a touch operation performed by a user, for example, after the user enters an App capable of performing the risk assessment and performs an operation of performing the risk assessment, the risk assessment device obtains the first request.
In response to the received first request, the risk assessment device outputs first interaction information. Wherein the first interactive information comprises a plurality of character analysis topics, and the character of the evaluation object can be analyzed based on the plurality of character analysis topics. The personality analysis questions can be subjective questions, the outputted personality analysis questions comprise question stems, the evaluation object answers questions of the question stems, and the answered contents are used as answers input into the risk evaluation equipment as evaluation objects; or, the personality analysis question can also be an objective question, the output personality analysis question not only comprises a question stem, but also comprises at least two alternative items, and the evaluation object selects answers from the at least two alternative items; or, the character analysis question is a subjective and objective combination question, the output character analysis question comprises a question stem and at least two optional items, and the evaluation object can select an answer from the at least two optional items and can input other answers.
In the related art, when risk assessment is performed, most of interaction information is output in a text form, and the defects of single output form, high writing degree and poor comprehension exist. In the embodiment of the application, the risk assessment equipment can output the first interactive information based on the multi-modal digital person, and the multi-modal digital person can output the first interactive information in a multi-dimensional form combining text, voice and limb actions by utilizing the speaking, writing and limb language expression capacity of the multi-modal digital person, so that the semantics and the context can be more highlighted, the attraction is improved, the evaluation object can understand the subject, and the evaluation duration is shortened. For example, in a virtual environment with multi-modal digital people, the first interaction information is output in a form of combining text and voice, the first interaction information can also be output in combination with video, and of course, the first interaction information can also be output in combination with other interaction modes.
In other embodiments, the risk assessment device may also obtain the first request from other terminals. At this time, the risk assessment device outputs the first interaction information to the other terminal, and the other terminal displays the first interaction information by using the multi-modal digital person.
Step S302, obtaining a first interaction result determined by the evaluation object based on the first interaction information.
The assessment object determines a first interaction result based on the output first interaction information, for example, the assessment object answers 10 personality analysis questions to obtain corresponding 10 answers, determines the 10 answers as the first interaction result, and inputs the first interaction result to the risk assessment device.
In the related art, when risk assessment is performed, most of the interactive information is output in a text form, and the assessment object selects a corresponding assessment option for the subject in the text form, so that the single text information input by the assessment object is still visible. In the embodiment of the application, when an evaluation object interacts with a multi-modal digital person, the multi-modal digital person obtains a first interaction result determined by the evaluation object for first interaction information by using listening and viewing capabilities of the multi-modal digital person, the first interaction result may be an answer given by the evaluation object according to text information "written" by the multi-modal digital person or "spoken" voice information, and the evaluation object may be a spoken answer or a corresponding answer selected by a risk assessment device. The risk assessment device determines to obtain a first interaction result based on the answer input by the assessment object.
Step S303, acquiring first expression information of the evaluation object when determining the first interaction result.
The risk assessment device acquires the expression of the assessment object when answering the character analysis question based on the multi-modal digital person, for example, the micro expression of the assessment object is acquired through an image acquisition device included in the risk assessment device.
In the related art, when risk assessment is performed, most of the interactive information is output in a text form, and the assessment object selects a corresponding assessment option for the subject in the text form, so that the single text information input by the assessment object is still visible. The risk assessment device cannot determine the first interaction result determined in which scene the assessment object is, and cannot determine the authenticity of the first interaction result only according to a single text message. In the embodiment of the application, when the evaluation object interacts with the multi-modal digital person, the multi-modal digital person obtains the micro-expression of the evaluation object when the evaluation object inputs the first interaction result by using the listening and viewing capabilities of the multi-modal digital person, namely, the image acquisition device acquires the image of the evaluation object including the expression characteristics in the answering process. And then extracting expressions based on the acquired images to obtain expressions of the evaluation object when the evaluation object answers each character analysis question, and determining first expression information based on the extracted expressions.
In this embodiment, the first expression information at least includes an eyeball sight line, a face orientation, an emotion, a blinking frequency, a mouth state, a face shield, and an eyebrow state. The first expression information may include state information of other parts than the facial state information, for example, limb movement information.
Step S304, determining character analysis data of the evaluation object based on the first interaction result and the first expression information.
The first interaction result is determined according to the first interaction information for determining the character analysis data, for example, the first interaction result is an evaluation result of the character analysis topic; the first expression information is information expressing emotion and sentiment, expressing thought emotion on the face or posture of the evaluation object, and the character of the evaluation object can be analyzed by combining the first expression information. And analyzing the acquired first interaction result and the first expression information by the risk assessment equipment to obtain character analysis data of the assessment object.
In the embodiment of the application, through interaction with an evaluation object in a multi-mode manner, based on an interaction result and expression information acquired in an interaction process, a personality label of the evaluation object is determined, such as whether the evaluation object has flexible thinking, whether things do have persistence, toughness and the like, whether the evaluation object is good at social contact and the like, and on the other hand, the personality of the evaluation object is determined. The personality analysis data of the evaluation object is determined through the first interaction result and the first expression information, the personality tendency of the evaluation object is judged from a psychological angle, the accuracy of the risk evaluation result can be improved, the risk of financial credit business is reduced, and bad account loss is reduced.
Step S305, risk analysis data is acquired.
Here, the step of acquiring risk analysis data may be performed at any time before determining the risk assessment result after the first request for performing risk assessment is acquired in step S301. And performing risk analysis according to objective data of the evaluation object, such as working conditions, income and expense conditions and the like of the evaluation object, so as to obtain risk analysis data. For example, if the work of the evaluation object is stable and the income and expense condition is stable, the risk of performing financial transaction with the evaluation object is small; whereas the risk is greater.
In some embodiments, obtaining risk analysis data may be implemented as: outputting second interactive information for determining risk analysis data, wherein the second interactive information comprises a plurality of risk analysis topics, and one risk analysis topic comprises a topic stem and/or a plurality of optional items; and acquiring a second interaction result determined by the evaluation object based on the second interaction information, and determining risk analysis data based on the second interaction result.
The mode of outputting the second interactive information may be the same as the mode of outputting the first interactive information, the second interactive information is a risk analysis topic, the topics are pre-stored in a storage space of the risk analysis device, the second interactive information is output when risk analysis is required, the evaluation object determines a second interactive result based on the second interactive information, and the second interactive result is input into the risk evaluation device, so that the risk evaluation device determines risk analysis data of the evaluation object based on the second interactive result.
Step S306, determining a risk assessment result based on the risk analysis data and the character analysis data.
In some embodiments, the risk analysis device obtains training data, and trains a preset risk assessment model to obtain a trained risk assessment model. And when a risk evaluation result is determined, inputting the risk analysis data and the character analysis data into the trained risk evaluation model to obtain a risk evaluation result. Here, the training data includes risk analysis training data and character analysis training data.
The risk analysis equipment considers the character of the evaluation object, and establishes a more complete and accurate user portrait by combining the risk analysis data of the evaluation object to assist in determining the risk evaluation result, so that more comprehensive, accurate and safe risk evaluation is given, and the bad account loss is effectively reduced.
The risk assessment method provided by the embodiment of the application comprises the following steps: responding to the acquired first request for risk assessment, and outputting first interaction information for determining character analysis data; acquiring a first interaction result determined by an evaluation object based on the first interaction information, and acquiring first expression information of the evaluation object when the first interaction result is determined; determining character analysis data of the evaluation object based on the first interaction result and the first expression information; acquiring risk analysis data, and determining a risk assessment result based on the risk analysis data and the personality analysis data. By analyzing the characters of the evaluation object and judging the character tendency of the evaluation object from the psychological aspect, the accuracy of the risk evaluation result can be improved, the risk of financial credit business can be reduced, and the bad account loss can be reduced.
Based on the foregoing embodiments, an embodiment of the present application further provides a risk assessment method, fig. 4 is a schematic flow chart of another implementation of the risk assessment method provided in the embodiment of the present application, and is applied to a risk assessment device in a network architecture shown in fig. 1A, as shown in fig. 4, the risk assessment method includes the following steps:
step S401, in response to the acquired first request for risk assessment, outputting first interaction information for determining the personality analysis data.
In the embodiment of the application, the risk assessment device may output the first interaction information based on the multi-modal digital person, for example, in a virtual environment with the multi-modal digital person, the first interaction information may be output in a form of combining text and voice, the first interaction information may also be output in combination with video, and of course, the first interaction information may also be output in combination with other interaction manners.
In other embodiments, the risk assessment device may also obtain the first request from other terminals. At this time, the risk assessment device outputs the first interaction information to the other terminal, and the other terminal outputs the first interaction information by using the multi-modal digital person.
Step S402, obtaining a first interaction result determined by the evaluation object based on the first interaction information.
Here, the first interactive information may include a plurality of personality analysis topics, and one personality analysis topic includes a topic stem and/or a plurality of alternatives.
In some embodiments, determining the first interaction result may be implemented as: obtaining each answering information corresponding to each character analysis question input by the evaluation object; and determining to obtain a first interaction result based on the answer information and a preset first scoring standard. Wherein the respective answer information is determined by the evaluation subject based on the respective personality analysis question including the question stem and/or the plurality of alternatives.
The evaluation object determines a first interaction result based on the output first interaction information, for example, the evaluation object answers 10 personality analysis questions to obtain corresponding 10 answer information, compares the 10 answer information with the first scoring standard to obtain 10 scores corresponding to the 10 answer information, and determines the sum of the 10 scores as the first interaction result.
Step S403, acquiring first expression information of the evaluation object when determining the first interaction result.
The risk assessment device acquires the expression of the assessment object when answering the character analysis question based on the multi-modal digital person, for example, the micro expression of the assessment object is acquired through an image acquisition device included in the risk assessment device.
When the evaluation object determines each answer information, carrying out image acquisition on the evaluation object to obtain a plurality of first face images; first expression information of the evaluation target is extracted based on the plurality of first face images. When the evaluation object interacts with the multi-mode digital person, the image acquisition device acquires an image of the evaluation object including expression features in the answering process. And then extracting expressions based on the acquired images to obtain expressions of the evaluation object when the evaluation object answers each character analysis question, and determining first expression information based on the extracted expressions.
In an embodiment of the application, the first expression information at least includes eye sight, face orientation, emotion, blink frequency, mouth state, face mask, and eyebrow state.
Step S404, determining character analysis data of the evaluation object based on the first interaction result and the first expression information.
In the embodiment of the application, through interaction with an evaluation object in a multi-mode manner, based on an interaction result and expression information acquired in an interaction process, a personality label of the evaluation object is determined, such as whether the evaluation object has flexible thinking, whether things do have persistence, toughness and the like, whether the evaluation object is good at social contact and the like, and on the other hand, the personality of the evaluation object is determined.
Step S405, determining first psychological analysis data of the evaluation target based on the first facial expression information.
The first expression information of the evaluation object in determining the first interaction result may be used to determine the character of the evaluation object, and may also be used to analyze the psychology of the evaluation object in determining the first interaction result.
Compared with expressions consciously made by people, the micro expression can reflect the real feeling and motivation of people. Although people ignore "micro-expressions," the human brain is still affected by them, changing the understanding of other expressions. So if a person naturally expresses a "happy" expression and does not contain a "micro expression", it can be concluded that the person is really happy. But if there is a 'smile' of 'smile' flashing there between, it is more likely that the 'happy' face is 'subtle' or 'untrustworthy', which is a false happy. Thus, the psychology of the evaluation subject can be analyzed based on the "micro expression".
For example, the assessment subjects answer the question "you are income per month: a.3000 or less; b.3000 to 8000; c.8000 to 20000; d.20000 or more. "the option selected immediately based on the first reaction is generally more trustworthy than the option selected after hesitation. In the hesitation process, the evaluation object may have a situation that the eyeball sight line is uncertain. For example, in psychology, psychological analysis is usually in recall when the eye looks down to the right, depending on the baseline change of each person; when the eye looks up to the left, psychological analysis is usually at a lie. Thereby, first psychological analysis data of the evaluation subject is determined using the first facial expression information based on psychology.
In some embodiments, determining the first mental analysis data of the evaluation object may be accomplished by: determining a second expression score value based on the first expression information and a preset third score standard; acquiring psychological analysis data corresponding to the second expression score value; and determining the acquired psychological analysis data as first psychological analysis data of the evaluation object.
Here, the third scoring criterion is a criterion based on psychological determination, which is stored in advance in the storage space of the risk assessment apparatus, and is a scoring criterion set based on an expression reference condition. And comparing the expression currently acquired by the evaluation object with the reference expression of the evaluation object, and determining the psychological characteristics of the evaluation object based on the comparison result.
In the embodiment of the application, the psychological characteristics of a plurality of expression analyses acquired by the multi-modal digital person can be fused, and the psychological analysis data of the evaluation object can be determined based on the fusion result, so that the accuracy of analysis psychology can be ensured.
Step S406, risk analysis data is acquired.
In some embodiments, obtaining risk analysis data may be implemented as: outputting second interactive information for determining risk analysis data, wherein the second interactive information comprises a plurality of risk analysis topics, and one risk analysis topic comprises a topic stem and/or a plurality of optional items; and acquiring a second interaction result determined by the evaluation object based on the second interaction information, and determining risk analysis data based on the second interaction result.
The mode of outputting the second interactive information may be the same as the mode of outputting the first interactive information, the second interactive information is a risk analysis topic, the topics are pre-stored in a storage space of the risk analysis device, the second interactive information is output when risk analysis is required, the evaluation object determines a second interactive result based on the second interactive information, and the second interactive result is input into the risk evaluation device, so that the risk evaluation device determines risk analysis data of the evaluation object based on the second interactive result.
Step S407, determining a risk assessment result based on the risk analysis data, the personality analysis data, and the first psychological analysis data.
In some embodiments, the risk analysis device obtains training data, and trains a preset risk assessment model to obtain a trained risk assessment model. And when a risk evaluation result is determined, inputting the risk analysis data, the character analysis data and the first psychological analysis data into a trained risk evaluation model to obtain a risk evaluation result. Here, the training data includes risk analysis training data, character analysis training data, and first psychoanalysis training data.
On the basis of the embodiment shown in fig. 3, the risk analysis device considers the psychology of the evaluation object in answering the character analysis data, and then combines the character analysis data and the risk analysis data of the evaluation object to establish a more complete and accurate user portrait to assist in determining the risk evaluation result, so as to give a more comprehensive, accurate and safe risk evaluation and further reduce bad account loss.
The risk assessment method provided by the embodiment of the application comprises the following steps: responding to the acquired first request for risk assessment, and outputting first interaction information for determining character analysis data; acquiring a first interaction result determined by an evaluation object based on the first interaction information, and acquiring first expression information of the evaluation object when the first interaction result is determined; determining character analysis data of the evaluation object based on the first interaction result and the first expression information; determining first psychological analysis data of the evaluation object based on the first expression information; acquiring risk analysis data, and determining a risk assessment result based on the risk analysis data and the personality analysis data. By carrying out character analysis and psychological analysis on the evaluation object and judging the character tendency and the psychological characteristics of the evaluation object from the psychological aspect, the accuracy of the risk evaluation result can be further improved, the risk of financial credit business is reduced, and the bad account loss is further reduced.
In some embodiments, the step S304 in the embodiment shown in fig. 3 or the step S404 "of determining the character analysis data of the evaluation object based on the first interaction result and the first expression information" in the embodiment shown in fig. 4 may be implemented by:
step S3041, determining a first expression score value based on the first expression information and a preset second score criterion.
The storage space of the risk assessment device stores a second scoring standard in advance, and the second scoring standard is set based on the expression reference condition of the assessment object. And scoring the first expression information by using a second scoring standard to obtain a first expression scoring value.
For example, when the evaluation subject answers 10 personality analysis questions, 20 first face images are collected, feature extraction is performed on the 20 first face images, 20 micro expressions are obtained through extraction, the 20 micro expressions are compared with expression benchmark conditions of the evaluation subject, the 20 micro expressions are scored based on a second scoring criterion, 20 scoring values are obtained, and the sum of the 20 scoring values can be determined as the first expression scoring value.
Step S3042, obtaining a first weight corresponding to the first interaction result and a second weight corresponding to the first expression score value.
The first weight and the second weight can be any value between 0 and 1, and are set according to the degree of importance, and if the interaction result is considered to be more important for character analysis, a larger value is set for the first weight; if the micro expression is deemed to be more important for the character analysis, a larger value is set for the second weight.
Step S3043, calculating a weighted sum of the first interaction result and the first expression score value based on the first weight and the second weight, and obtaining a weighted value.
Recording a first interaction result as P, recording a first expression score value as Q, and recording a first weight as w1The second weight is w2Then, the weight value a is expressed by formula (1):
A＝w1*P+w2*Q (1)；
step S3044, obtaining character analysis data corresponding to the weighted value.
The risk assessment device stores a corresponding relation table of a numerical range and character analysis data, for example, a weighted value a has a value range of 0 to 30, a weighted value a has a value range of 0 to 10 corresponding to character analysis data of a first character (e.g., inward), a weighted value a has a value range of 11 to 20 corresponding to character analysis data of a second character (e.g., neutral), and a weighted value a has a value range of 21 to 30 corresponding to character analysis data of a third character (e.g., outward), and the relation table is queried according to the weighted value to obtain character analysis data corresponding to the weighted value, for example, a has a value of 20, and obtain character analysis data of neutral characters.
In step S3045, the acquired character analysis data is determined as the character analysis data of the evaluation target.
In the embodiment of the application, a weighted value is determined based on the first interaction result and the first expression information, and the character analysis data of the evaluation object is obtained by querying a pre-stored relationship table according to the weighted value, so that the character analysis of the evaluation object is realized. When the character analysis is carried out, the character test questions are considered, the expression information of the evaluation object is obtained when the questions are answered, and the accuracy of character analysis data can be improved, so that the accuracy of a risk evaluation result is further improved, and the risk of financial credit business is reduced.
In some embodiments, the step S305 in the embodiment shown in fig. 3 or the step S406 "acquiring risk analysis data" in the embodiment shown in fig. 4 may be implemented by the following steps:
and S3051, outputting second interaction information for determining risk analysis data.
Here, the second interactive information includes a plurality of risk analysis topics, and one risk analysis topic includes a topic stem and/or a plurality of alternatives.
In the embodiment of the application, the risk assessment apparatus may output the second interaction information based on the multi-modal digital person, for example, in a virtual environment with the multi-modal digital person, the second interaction information may be output in a form of combining text and voice, may also be output in combination with video, and of course, may also be output in combination with other interaction manners.
Step S3052, a second interaction result determined by the evaluation object based on the second interaction information is obtained.
The evaluation object determines a second interaction result based on the output second interaction information, for example, the evaluation object answers 10 risk analysis questions to obtain corresponding 10 answer information, compares the 10 answer information with a preset scoring standard to obtain 10 scores corresponding to the 10 answer information, and determines the sum of the 10 scores as the second interaction result.
In the embodiment of the application, when the evaluation object interacts with the multi-modal digital person, the multi-modal digital person obtains a second interaction result determined by the evaluation object for second interaction information by using listening and viewing capabilities of the multi-modal digital person, the second interaction result may be an answer given by the evaluation object according to text information "written" by the multi-modal digital person or "spoken" voice information, and the evaluation object may be a spoken answer or a corresponding answer selected by the risk assessment device. And the risk assessment equipment determines to obtain a second interaction result based on the answer input by the assessment object.
And S3053, determining risk analysis data based on the second interaction result.
The risk assessment device stores a corresponding relation table of a numerical range and risk analysis data, for example, the value range of the second interaction result is 0 to 10, 0 to 2 corresponds to low-risk analysis data, 3 to 7 corresponds to medium-risk analysis data, 8 to 10 corresponds to high-risk analysis data, and the relation table is queried according to the second interaction result to obtain corresponding risk analysis data.
In the embodiment of the application, after the second interaction result is determined according to the second interaction information, the pre-stored relation table is inquired to obtain the risk analysis data of the evaluation object, and the risk analysis of the evaluation object is realized.
Further, after the step S3053, the method may further include the steps of:
step S3054, second expression information of the evaluation object when the second interaction result is determined is obtained.
In the embodiment of the application, when the evaluation object interacts with the multi-modal digital person, the multi-modal digital person obtains the micro-expression when the evaluation object inputs the second interaction result by using the listening and viewing capabilities of the multi-modal digital person, namely, the image acquisition device acquires the image of the evaluation object including the expression characteristics in the answering process. And then extracting expressions based on the acquired images to obtain expressions of the evaluation object when the evaluation object answers each risk analysis topic, and determining second expression information based on the extracted expressions. The second expression information at least comprises eyeball sight, face orientation, emotion, blink frequency, mouth state, face shielding and eyebrow state.
In the embodiment of the application, the second expression information is different from the first expression information in that the first expression information corresponds to an expression of an evaluation object when answering a character analysis question, the second expression information corresponds to an expression of the evaluation object when answering a risk analysis question, and the second expression information and the first expression information are acquired in the same manner.
Step S3055, determining second psychological analysis data of the evaluation object based on the second expression information.
Here, the implementation of determining the second psychological analysis data of the evaluation target based on the second facial expression information is the same as the implementation of determining the first psychological analysis data of the evaluation target based on the first facial expression information, specifically referring to step S405.
Correspondingly, after step S3055, step S306 in the embodiment shown in fig. 3 is replaced by:
step S3056, determining a risk assessment result based on the risk analysis data, the second psychological analysis data and the character analysis data.
On the basis of the embodiment shown in fig. 3, the risk analysis device considers the psychology of the evaluation object when answering the risk analysis data, and then combines the character analysis data and the risk analysis data of the evaluation object to establish a more complete and accurate user portrait to assist in determining the risk evaluation result, so as to give a more comprehensive, accurate and safe risk evaluation and further reduce bad account loss.
Based on the foregoing embodiments, a risk assessment method is further provided in an embodiment of the present application, fig. 5 is a schematic flow chart of another implementation of the risk assessment method provided in the embodiment of the present application, and is applied to a risk assessment device in a network architecture shown in fig. 1A, as shown in fig. 5, the risk assessment method includes the following steps:
step S501, a first request is acquired.
Here, the risk assessment device may obtain the first request based on a touch operation performed by the user, for example, after the user enters an App capable of performing risk assessment and performs an operation of performing risk assessment, the risk assessment device obtains the first request.
Step S502, responding to the first request, and outputting first interaction information for determining character analysis data.
Here, the first interactive information includes a plurality of personality analysis topics, and one personality analysis topic includes a topic stem and/or a plurality of alternatives. In the embodiment of the application, the risk assessment device can output the first interaction information based on the multi-modal digital person.
Step S503, obtaining each answer information corresponding to each character analysis question inputted by the evaluation object.
Here, the respective answer information is determined by the evaluation subject based on the respective character analysis questions including the question stem and/or the plurality of alternatives.
Step S504, determining to obtain a first interaction result based on the answer information and a preset first scoring criterion.
The storage space of the risk assessment device stores a first scoring standard in advance, and the first scoring standard is set based on psychology. And scoring each answering information by using a first scoring standard to obtain a first interaction result.
And step S505, when the evaluation object determines each answer information, carrying out image acquisition on the evaluation object to obtain a plurality of first face images.
The risk assessment device obtains a plurality of first facial images of an assessment subject in response to a personality analysis topic, e.g., one facial image for each personality analysis topic, based on a multimodal digital person. Then, a micro-expression of the evaluation object is extracted from the plurality of first face images.
Step S506, extracting first expression information of the evaluation target based on the plurality of first face images.
Here, the first expression information includes at least an eyeball sight line, a face orientation, an emotion, a blinking frequency, a mouth state, a face mask, and an eyebrow state.
Step S507, determining a first expression score value based on the first expression information and a preset second score criterion.
The storage space of the risk assessment device stores a second scoring standard in advance, and the second scoring standard is set based on the expression reference condition of the assessment object. And scoring the first expression information by using a second scoring standard to obtain a first expression scoring value. The second scoring criterion here is to test the character characteristics of the evaluation subject based on the expression information.
Step S508, a first weight corresponding to the first interaction result and a second weight corresponding to the first expression score value are obtained.
The first weight and the second weight can be set to any value between 0 and 1 according to the degree of importance, and if the interaction result is considered to be more important for the character analysis, a larger value is set for the first weight; if the micro expression is deemed to be more important for the character analysis, a larger value is set for the second weight.
Step S509, calculating a weighted sum of the first interaction result and the first expression score value based on the first weight and the second weight, and obtaining a weighted value.
Step S510, obtaining the character analysis data corresponding to the weighted value, and determining the obtained character analysis data as the character analysis data of the evaluation target.
The risk assessment equipment stores a corresponding relation table of the numerical range and the character analysis data, and inquires the relation table according to the weighted value to obtain the character analysis data corresponding to the weighted value.
Step S511, determining a second expression score value based on the first expression information and a preset third score criterion.
The storage space of the risk assessment device stores a third scoring standard in advance, and the third scoring standard is set based on the expression reference condition of the assessment object. And scoring the first expression information by using a third scoring standard to obtain a second expression scoring value. The third scoring criterion here is to test psychological characteristics of the assessment subject based on the expression information.
Step S512, obtaining psychological analysis data corresponding to the second expression score value.
The risk assessment device stores a corresponding relation table of the numerical range and the psychological analysis data, and queries the relation table according to the second expression score value to obtain the psychological analysis data corresponding to the second expression score value.
In step S513, the acquired psychological analysis data is determined as the first psychological analysis data of the evaluation target.
And step S514, outputting second interaction information for determining risk analysis data.
Here, the second interactive information includes a plurality of risk analysis topics, and one risk analysis topic includes a topic stem and/or a plurality of alternatives.
Step S515, obtaining a second interaction result determined by the evaluation object based on the second interaction information.
The storage space of the risk assessment device stores a fourth scoring criterion in advance, and the fourth scoring criterion is set based on risk assessment experience. And scoring the answer information of each risk analysis question by using a fourth scoring standard to obtain a second interaction result.
Step S516, determining risk analysis data based on the second interaction result.
The risk evaluation equipment stores a corresponding relation table of the numerical range and the risk analysis data, and inquires the relation table according to the second interaction result to obtain the risk analysis data corresponding to the second interaction result.
Step S517, second expression information of the evaluation object when the second interaction result is determined is obtained.
When the evaluation object determines the answer information of each risk analysis question, image acquisition is carried out on the evaluation object to obtain a plurality of second face images; second expression information of the evaluation target is then extracted based on the plurality of second face images.
Here, the second expression information includes at least an eyeball sight line, a face orientation, an emotion, a blinking frequency, a mouth state, a face mask, and an eyebrow state.
Step S518, determining second psychological analysis data of the evaluation target based on the second facial expression information.
Step S519, determining a risk assessment result based on the risk analysis data, the second psychological analysis data, the character analysis data, and the first psychological analysis data.
In some embodiments, the risk analysis device obtains training data, and trains a preset risk assessment model to obtain a trained risk assessment model. And when a risk evaluation result is determined, inputting the risk analysis data, the second psychological analysis data, the character analysis data and the first psychological analysis data into a trained risk evaluation model to obtain a risk evaluation result. Here, the training data includes risk analysis training data, second psychoanalysis training data, character analysis training data, and first psychoanalysis training data.
According to the risk assessment method provided by the embodiment of the application, the evaluation object is subjected to personality analysis by interacting the risk assessment equipment with the evaluation object, the personality tendency of the evaluation object is judged from a psychological perspective, the first psychological characteristic of the evaluation object during the personality analysis and the second psychological characteristic of the evaluation object during the risk analysis are analyzed, and the personality tendency, the psychological characteristic and the risk analysis of the evaluation object are combined, so that the accuracy of a risk assessment result can be improved, the risk of financial credit business is reduced, and the bad account loss is reduced.
Based on the foregoing embodiments, an embodiment of the present application further provides a risk assessment method, and fig. 6 is a schematic flow chart of another implementation of the risk assessment method provided in the embodiment of the present application, which is applied to the network architecture shown in fig. 1B, as shown in fig. 6, the risk assessment method includes the following steps:
step S601, the terminal sends a first request to the risk assessment device.
Here, the first request is for risk assessment. The method comprises the steps that an App capable of performing risk assessment is installed on a terminal, and after the App performs operation of performing risk assessment, the terminal is triggered to send a first request to risk assessment equipment.
Step S602, the risk assessment device responds to the first request, and outputs first interaction information for determining the character analysis data to the terminal.
The risk assessment equipment stores a plurality of questions for character analysis in advance, and after a first request is obtained, first interactive information carrying the plurality of characters analysis questions is sent to the terminal.
Step S603, the terminal receives a first interaction result determined and input by the evaluation object based on the first interaction information.
And the evaluation object carries out character test based on a plurality of character analysis questions output by the terminal to obtain a first interaction result.
Step S604, the terminal sends the first interaction result to the risk assessment device.
Here, the interaction information in the interaction process between the terminal and the risk assessment device carries identification information capable of determining the identity of the assessment object, so that the risk assessment device knows which assessment object to perform risk assessment.
And step S605, when the terminal determines each answer information of the evaluation object, acquiring images of the evaluation object to obtain a plurality of first face images.
In step S606, the terminal determines first expression information based on the plurality of first face images.
In some embodiments, the terminal may determine first expression information of the assessment target based on the multimodal digital person.
Step S607, the terminal sends the first expression information to risk assessment equipment.
Step S608, the risk assessment apparatus determines the personality analysis data of the assessment object based on the first interaction result and the first expression information.
And the risk assessment equipment determines a weighted value based on the first interaction result and the first expression information, and queries a relation table of a numerical range and character analysis data stored in the storage space according to the weighted value to obtain character analysis data corresponding to the weighted value.
In step S609, the risk assessment device determines first psychological analysis data of the assessment target based on the first expression information.
The risk assessment equipment determines a score value based on the first expression information, and queries a relation table of a numerical range and psychological analysis data stored in a storage space according to the score value to obtain first psychological analysis data corresponding to the score value. The first psychological analysis data herein is a psychological characteristic when the evaluation target performs character analysis.
Step S610, the risk assessment device outputs second interaction information for determining risk analysis data to the terminal.
Here, the second interactive information includes a plurality of risk analysis topics, and one risk analysis topic includes a topic stem and/or a plurality of alternatives.
Step S611, the terminal receives a second interaction result determined and input by the evaluation object based on the second interaction information.
And the evaluation object carries out risk analysis test based on the plurality of risk analysis topics output by the terminal to obtain a second interaction result.
And step S612, the terminal sends the second interaction result to the risk assessment equipment.
Step S613, when the terminal determines the second interaction result, performing image acquisition on the evaluation object to obtain a plurality of second face images.
In step S614, the terminal determines second expression information based on the plurality of second face images.
In some embodiments, the terminal may determine second expression information of the evaluation object based on the multi-modal digital person.
And step S615, the terminal sends the second expression information to the risk assessment equipment.
In step S616, the risk assessment apparatus determines second psychological analysis data of the assessment target based on the second expression information.
And the risk assessment equipment determines a score value based on the second expression information, and queries a relation table of the numerical range and the psychological analysis data stored in the storage space according to the score value to obtain second psychological analysis data corresponding to the score value. The second psychological analysis data herein is a psychological characteristic when risk analysis is performed on the evaluation target.
In step S617, the risk assessment apparatus determines risk analysis data based on the second interaction information.
The risk evaluation equipment stores a corresponding relation table of the numerical range and the risk analysis data, and inquires the relation table according to the second interaction result to obtain the risk analysis data corresponding to the second interaction result.
In step S618, the risk assessment device determines a risk assessment result based on the risk analysis data, the second psychological analysis data, the character analysis data, and the first psychological analysis data.
In some embodiments, the risk analysis device obtains training data, and trains a preset risk assessment model to obtain a trained risk assessment model. And when a risk evaluation result is determined, inputting the risk analysis data, the second psychological analysis data, the character analysis data and the first psychological analysis data into a trained risk evaluation model to obtain a risk evaluation result. Here, the training data includes risk analysis training data, second psychoanalysis training data, character analysis training data, and first psychoanalysis training data.
According to the risk assessment method provided by the embodiment of the application, the risk assessment equipment interacts with the assessment object based on the terminal, performs character analysis on the assessment object, judges the character tendency of the assessment object from a psychological perspective, analyzes the first psychological characteristic of the assessment object during character analysis and the second psychological characteristic of the assessment object during risk analysis by using the face image of the assessment object acquired by the terminal, and can improve the accuracy of a risk assessment result by combining the character tendency, the psychological characteristic and the risk analysis of the assessment object, thereby being beneficial to reducing the risk of financial credit business and further reducing bad account loss.
Next, an exemplary application of the embodiment of the present application in a practical application scenario will be described.
In the current financial wind control model, for loan individual users and enterprise users, the wind control model is mainly used for risk assessment according to user objective data (such as user personal specific information, all related data in other financial systems, enterprise specific information, industry development prospects and the like). The risk assessment only considers the objective information of the user, and lacks the risk assessment on subjective influence factors such as the psychology and the character of the user, so that the bad account conditions after loan fraud are more, and the bad account rate is higher.
In order to solve the above problems, an embodiment of the present application provides a robot for detecting subjective psychological characteristics of loan users by using multi-modal digital human-technology assisted risk assessment, and fig. 7 is a schematic flow chart of another implementation of the risk assessment provided by the embodiment of the present application, as shown in fig. 7, the method includes the following steps:
step S701, the user accesses to the financial institution multi-modal digital person.
Before the user makes a loan, the user can access the multi-mode digital person through the APP of the financial institution. I.e. the user registers and logs in the financial APP of the connected multimodal digital person.
Step S702, the multi-modal digital person interacts with the loan user according to a series of professional psychological assessment questions.
Here, the evaluation questions are determined by professional evaluation institutions, and are generally selection questions. The multi-modal digital person can display evaluation questions and options through a display interface of the terminal, and the options are manually selected by a user; or the multi-modal digital person can play the evaluation questions and options through a loudspeaker of the terminal, a microphone of the terminal collects the voice of the user, and the voice is recognized to determine the options selected by the user; of course, interaction may be performed in other ways.
And step S703, carrying out recognition analysis on the micro expression during user interaction through face recognition.
In the interaction process, the multi-modal digital person acquires images of a user through a camera of the terminal or an external image acquisition device which can transmit information with the multi-modal digital person, identifies and analyzes the acquired images, and extracts micro expressions and actions in the user interaction process, such as thought emotions expressed on faces (sight lines, corners of mouths, eyebrows and the like) or behavior gestures.
And step S704, evaluating the personality tendency and the psychological characteristics of the user to obtain the psychological characteristic data of the user.
By means of analysis of a multi-mode digital psychology angle, the method evaluates the psychology intrinsic characteristics such as the personality tendency of loan users, for example, whether CEO of small and micro enterprises has flexible thinking, whether things do have persistence, toughness and the like, and further estimates the success of the operation of the enterprises from the other side.
And step S705, combining the acquired user psychological characteristic data with the client data in the wind control model, and performing risk assessment on the user to obtain an assessment result.
User psychological characteristic data acquired by the multi-modal digital person, namely loan user character psychological analysis data, is added into the wind control model, and a more perfect and accurate loan user portrait is established to assist the wind control model to give more comprehensive, accurate and safe risk assessment, so that bad account rate is effectively reduced.
According to the risk assessment method provided by the embodiment of the application, the face recognition technology is utilized to analyze the intrinsic subjective psychological tendency of the loan user such as fine expressions and the like, the personality tendency and the psychological characteristics of the loan user are judged from the psychological perspective, the personality tendency and the psychological characteristics are added into the wind control model, and the risk assessment is carried out by combining the existing objective data in the wind control model, so that the risk assessment result is more comprehensive, the accuracy and the safety are higher, the risk avoidance can be realized, and the bad account rate is reduced.
Continuing with the exemplary structure of the risk assessment device provided by the embodiments of the present application implemented as software modules, in some embodiments, as shown in fig. 2, the risk assessment device 80 stored in the memory 140 is applied to perform risk assessment, and the software modules in the risk assessment device 80 may include:
the output module 81 is configured to output first interaction information used for determining the personality analysis data in response to the acquired first request for risk assessment;
a first obtaining module 82, configured to obtain a first interaction result determined by the evaluation object based on the first interaction information;
a second obtaining module 83, configured to obtain first expression information of the evaluation object when determining the first interaction result;
a first determination module 84, configured to determine character analysis data of the evaluation object based on the first interaction result and the first expression information;
a third obtaining module 85, configured to obtain risk analysis data;
a second determination module 86 for determining a risk assessment result based on the risk analysis data and the personality analysis data.
In some embodiments, the risk assessment apparatus 80 may further include:
a third determination module, configured to determine first psychological analysis data of the evaluation object based on the first expression information;
accordingly, the second determining module 86 is further configured to:
determining a risk assessment result based on the risk analysis data, the personality analysis data, and the first mental analysis data.
In some embodiments, the first interactive information comprises a plurality of personality analysis topics, one personality analysis topic comprising a stem and/or a plurality of alternatives;
the first obtaining module 82 is further configured to:
obtaining each answer information corresponding to each character analysis question input by the evaluation object, wherein each answer information is determined by the evaluation object based on each character analysis question including a question stem and/or a plurality of optional items;
and determining to obtain a first interaction result based on the answer information and a preset first scoring standard.
In some embodiments, the second obtaining module 83 is further configured to:
when the evaluation object determines each answer information, carrying out image acquisition on the evaluation object to obtain a plurality of first face images;
extracting first expression information of the evaluation object based on the plurality of first face images, wherein the first expression information at least comprises eyeball sight lines, face orientation, emotion, blink frequency, mouth states, face occlusion and eyebrow states.
In some embodiments, the first determining module 84 is further configured to:
determining a first expression score value based on the first expression information and a preset second score standard;
acquiring a first weight corresponding to the first interaction result and a second weight corresponding to the first expression score value;
calculating a weighted sum of the first interaction result and the first expression score value based on the first weight and the second weight to obtain a weighted value;
and acquiring character analysis data corresponding to the weighted value, and determining the acquired character analysis data as character analysis data of the evaluation object.
In some embodiments, the third determining module is further configured to:
determining a second expression score value based on the first expression information and a preset third score standard;
and acquiring psychological analysis data corresponding to the second expression score value, and determining the acquired psychological analysis data as first psychological analysis data of the evaluation object.
In some embodiments, the third obtaining module 85 is further configured to:
outputting second interactive information for determining risk analysis data, wherein the second interactive information comprises a plurality of risk analysis topics, and one risk analysis topic comprises a topic stem and/or a plurality of optional items;
and acquiring a second interaction result determined by the evaluation object based on the second interaction information, and determining risk analysis data based on the second interaction result.
In some embodiments, the risk assessment apparatus 80 may further include:
the fourth obtaining module is used for obtaining second expression information of the evaluation object when the second interaction result is determined;
a fourth determination module, configured to determine second psychological analysis data of the evaluation object based on the second expression information;
accordingly, the second determining module 86 is further configured to:
determining a risk assessment result based on the risk analysis data, the second mental analysis data, and the personality analysis data.
Here, it should be noted that: the above description of the embodiment of the risk assessment device is similar to the above description of the method, and has the same beneficial effects as the embodiment of the method. For technical details not disclosed in the embodiments of the risk assessment device of the present application, a person skilled in the art shall refer to the description of the embodiments of the method of the present application for understanding.
Embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the risk assessment method described in the embodiment of the present application.
Embodiments of the present application provide a storage medium having stored therein executable instructions, which when executed by a processor, will cause the processor to perform the methods provided by embodiments of the present application, for example, the methods as illustrated in fig. 3 to 7.
In some embodiments, the storage medium may be memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or may be various devices including one or any combination of the above memories.
In some embodiments, executable instructions may be written in any form of programming language (including compiled or interpreted languages), in the form of programs, software modules, scripts or code, and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.
By way of example, executable instructions may correspond, but do not necessarily have to correspond, to files in a file system, and may be stored in a portion of a file that holds other programs or data, such as in one or more scripts in a hypertext Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub-programs, or portions of code).
By way of example, executable instructions may be deployed to be executed on one computing device or on multiple computing devices at one site or distributed across multiple sites and interconnected by a communication network.
The above description is only an example of the present application, and is not intended to limit the scope of the present application. Any modification, equivalent replacement, and improvement made within the spirit and scope of the present application are included in the protection scope of the present application.

Claims (12)

1. A method of risk assessment, the method comprising:
responding to the acquired first request for risk assessment, and outputting first interaction information for determining character analysis data;
acquiring a first interaction result determined by an evaluation object based on the first interaction information, and acquiring first expression information of the evaluation object when the first interaction result is determined;
determining character analysis data of the evaluation object based on the first interaction result and the first expression information;
acquiring risk analysis data, and determining a risk assessment result based on the risk analysis data and the personality analysis data.
2. The method of claim 1, further comprising:
determining first psychological analysis data of the evaluation object based on the first expression information;
accordingly, the determining a risk assessment result based on the risk analysis data and the personality analysis data includes:
determining a risk assessment result based on the risk analysis data, the personality analysis data, and the first mental analysis data.
3. The method according to claim 1 or 2, wherein the first interactive information comprises a plurality of personality analysis topics, and one personality analysis topic comprises a question stem and/or a plurality of options;
the obtaining of the first interaction result determined by the evaluation object based on the first interaction information includes:
obtaining each answer information corresponding to each character analysis question input by the evaluation object, wherein each answer information is determined by the evaluation object based on each character analysis question including a question stem and/or a plurality of optional items;
and determining to obtain a first interaction result based on the answer information and a preset first scoring standard.
4. The method of claim 3, wherein the obtaining of the first expression information of the evaluation object in determining the first interaction result comprises:
when the evaluation object determines each answer information, carrying out image acquisition on the evaluation object to obtain a plurality of first face images;
extracting first expression information of the evaluation object based on the plurality of first face images, wherein the first expression information at least comprises eyeball sight lines, face orientation, emotion, blink frequency, mouth states, face occlusion and eyebrow states.
5. The method of claim 1 or 2, wherein determining personality analysis data for the assessment object based on the first interaction result and the first expression information comprises:
determining a first expression score value based on the first expression information and a preset second score standard;
acquiring a first weight corresponding to the first interaction result and a second weight corresponding to the first expression score value;
calculating a weighted sum of the first interaction result and the first expression score value based on the first weight and the second weight to obtain a weighted value;
and acquiring character analysis data corresponding to the weighted value, and determining the acquired character analysis data as character analysis data of the evaluation object.
6. The method of claim 2, wherein determining the first psychoanalytic data of the assessment target based on the first expression information comprises:
determining a second expression score value based on the first expression information and a preset third score standard;
and acquiring psychological analysis data corresponding to the second expression score value, and determining the acquired psychological analysis data as first psychological analysis data of the evaluation object.
7. The method of claim 1 or 2, wherein said obtaining risk analysis data comprises:
outputting second interactive information for determining risk analysis data, wherein the second interactive information comprises a plurality of risk analysis topics, and one risk analysis topic comprises a topic stem and/or a plurality of optional items;
and acquiring a second interaction result determined by the evaluation object based on the second interaction information, and determining risk analysis data based on the second interaction result.
8. The method of claim 7, further comprising:
acquiring second expression information of the evaluation object when the second interaction result is determined;
determining second psychological analysis data of the evaluation object based on the second facial expression information;
accordingly, the determining a risk assessment result based on the risk analysis data and the personality analysis data includes:
determining a risk assessment result based on the risk analysis data, the second mental analysis data, and the personality analysis data.
9. A risk assessment device, characterized in that the device comprises:
the output module is used for responding to the acquired first request for risk assessment and outputting first interaction information for determining character analysis data;
the first acquisition module is used for acquiring a first interaction result determined by the evaluation object based on the first interaction information;
the second acquisition module is used for acquiring first expression information of the evaluation object when the first interaction result is determined;
a first determination module, configured to determine personality analysis data of the evaluation object based on the first interaction result and the first expression information;
the third acquisition module is used for acquiring risk analysis data;
a second determination module to determine a risk assessment result based on the risk analysis data and the personality analysis data.
10. A risk assessment device, characterized in that the device comprises:
a memory for storing executable instructions;
a processor for implementing the method of any one of claims 1 to 8 when executing executable instructions stored in the memory.
11. A computer-readable storage medium having stored thereon executable instructions for causing a processor, when executed, to implement the method of any one of claims 1 to 8.
12. A computer program product comprising a computer program, characterized in that the computer program realizes the method of any of claims 1 to 8 when executed by a processor.
CN202110209854.0A 2021-02-24 2021-02-24 Risk assessment method, apparatus, computer-readable storage medium, and program product Pending CN112819609A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110209854.0A CN112819609A (en) 2021-02-24 2021-02-24 Risk assessment method, apparatus, computer-readable storage medium, and program product

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110209854.0A CN112819609A (en) 2021-02-24 2021-02-24 Risk assessment method, apparatus, computer-readable storage medium, and program product

Publications (1)

Publication Number Publication Date
CN112819609A true CN112819609A (en) 2021-05-18

Family

ID=75865577

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110209854.0A Pending CN112819609A (en) 2021-02-24 2021-02-24 Risk assessment method, apparatus, computer-readable storage medium, and program product

Country Status (1)

Country Link
CN (1) CN112819609A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117033724A (en) * 2023-08-24 2023-11-10 青海昇云信息科技有限公司 Multi-mode data retrieval method based on semantic association

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109472206A (en) * 2018-10-11 2019-03-15 平安科技（深圳）有限公司 Methods of risk assessment, device, equipment and medium based on micro- expression
WO2019074446A1 (en) * 2017-10-13 2019-04-18 Digital Alpha Group Pte Ltd System and method for processing a loan application
CN109829358A (en) * 2018-12-14 2019-05-31 深圳壹账通智能科技有限公司 Micro- expression loan control method, device, computer equipment and storage medium
CN109872026A (en) * 2018-12-14 2019-06-11 深圳壹账通智能科技有限公司 Evaluation result generation method, device, equipment and computer readable storage medium
WO2019153486A1 (en) * 2018-02-07 2019-08-15 平安科技（深圳）有限公司 Customer risk assessment method, apparatus and device, and computer-readable storage medium
CN110210380A (en) * 2019-05-30 2019-09-06 盐城工学院 The analysis method of personality is generated based on Expression Recognition and psychology test
CN110569347A (en) * 2019-09-10 2019-12-13 出门问问信息科技有限公司 Data processing method and device, storage medium and electronic equipment

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019074446A1 (en) * 2017-10-13 2019-04-18 Digital Alpha Group Pte Ltd System and method for processing a loan application
WO2019153486A1 (en) * 2018-02-07 2019-08-15 平安科技（深圳）有限公司 Customer risk assessment method, apparatus and device, and computer-readable storage medium
CN109472206A (en) * 2018-10-11 2019-03-15 平安科技（深圳）有限公司 Methods of risk assessment, device, equipment and medium based on micro- expression
CN109829358A (en) * 2018-12-14 2019-05-31 深圳壹账通智能科技有限公司 Micro- expression loan control method, device, computer equipment and storage medium
CN109872026A (en) * 2018-12-14 2019-06-11 深圳壹账通智能科技有限公司 Evaluation result generation method, device, equipment and computer readable storage medium
CN110210380A (en) * 2019-05-30 2019-09-06 盐城工学院 The analysis method of personality is generated based on Expression Recognition and psychology test
CN110569347A (en) * 2019-09-10 2019-12-13 出门问问信息科技有限公司 Data processing method and device, storage medium and electronic equipment

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117033724A (en) * 2023-08-24 2023-11-10 青海昇云信息科技有限公司 Multi-mode data retrieval method based on semantic association

Similar Documents

Publication Publication Date Title
Macdonald et al. Gaze in a real-world social interaction: A dual eye-tracking study
Baylor et al. A qualitative study of interference with communicative participation across communication disorders in adults
US20200082928A1 (en) Assisting psychological cure in automated chatting
US20130266925A1 (en) Embedded Conversational Agent-Based Kiosk for Automated Interviewing
KR102118585B1 (en) Smart Mirror Chatbot System and Method for Senior Care
CN109801105A (en) Service methods of marking, device, equipment and storage medium based on artificial intelligence
US10896688B2 (en) Real-time conversation analysis system
CN109711200A (en) Accurate poverty alleviation method, apparatus, equipment and medium based on block chain
CN111542833A (en) Classification screening engine for document authentication
US20210271864A1 (en) Applying multi-channel communication metrics and semantic analysis to human interaction data extraction
US20230080660A1 (en) Systems and method for visual-audio processing for real-time feedback
CN110047518A (en) A kind of speech emotional analysis system
Mihoub et al. Learning multimodal behavioral models for face-to-face social interaction
CN116762125A (en) Environment collaboration intelligent system and method
CN115034886A (en) Default risk prediction method and device
CN112819609A (en) Risk assessment method, apparatus, computer-readable storage medium, and program product
CN107491757A (en) A kind of antifraud system and control method based on living body characteristics
CN111027358A (en) Dictation and reading method based on writing progress and electronic equipment
CN110782119A (en) Insurance agent selection method, device and equipment based on artificial intelligence
EP4064031A1 (en) Method and system for tracking in extended reality using voice commmand
US20180232625A1 (en) Avatar relationships
Battina et al. Innovative study of an AI voice based smart Device to assist deaf people in understanding and responding to their body language
Dokania et al. An assistive interface protocol for communication between visually and hearing-speech impaired persons in internet platform
CN112991076A (en) Information processing method and device
Mehigan et al. Modelling an holistic artificial intelligent education model for optimal learner engagement and inclusion

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination