Risk assessment method and device based on user psychological assessment data

Abstract

The invention discloses a risk assessment method and device based on user psychological assessment data, and the method and device can timely obtain a current feedback data set for psychological assessment of a student user, an emotion recognition result corresponding to a test process video data set, a current student user course data set, and a current student user test score data set. And multi-dimensional data of the student user is effectively combined for classification and evaluation, so that a current classification output result which is high in real-time performance and is more visually and quantitatively displayed is obtained, and a teacher user of the psychological health major can view and analyze the result.

Description

Technical Field

[0001] The present invention relates to the field of intelligent decision-making technology of artificial intelligence, and in particular to a risk assessment method and device based on user psychological assessment data. Background Art

[0002] Psychological testing is a process of quantifying and analyzing people's cognitive, behavioral, emotional and other psychological activities according to certain laws and psychological principles and using certain operating procedures. At present, when conducting psychological tests on students in school, the mental health scale is usually used to evaluate the students' mental state. The mental health scale is divided into different categories such as symptom orientation, adaptation orientation and psychological quality orientation. However, when using the mental health scale to test students, the test content is often fixed, and it is impossible to recognize the emotions of students' expressions during the test. Therefore, there are limitations such as one-sidedness and singleness for different test objects. In addition, the psychological testing method using the mental health scale belongs to the traditional manual method, which not only has low efficiency in test data processing, but also cannot guarantee the effectiveness of test results, resulting in low accuracy of test results. In order to solve the above problems, it is necessary to provide a more effective risk assessment method based on user psychological test data. Summary of the invention

[0003] The embodiment of the present invention provides a risk assessment method and device based on user psychological assessment data, aiming to solve the problem that when students are tested using a mental health scale in the prior art, not only is it difficult to ensure the effectiveness and efficiency of the answer results to the test questions, but also manual evaluation is required after the results are obtained, resulting in low data processing efficiency.

[0004] In a first aspect, an embodiment of the present invention provides a risk assessment method based on user psychological assessment data, which includes: In response to the evaluation instruction, obtaining the student user information of the student object to be evaluated in the evaluation instruction; wherein the student user information at least includes the student user unique identification number, the student user tag and the student historical evaluation result; Based on a pre-built psychological assessment knowledge base, the student user information and a preset test data generation strategy, a current test data set is generated, and the current test data set is sent to a receiving terminal corresponding to the student user information; Receiving a current feedback data set sent by the receiving terminal and corresponding to the current test data set, and acquiring a first current classification result based on the current feedback data set and a pre-trained first classification model; Receiving a test process video data set sent by the receiving terminal and corresponding to the current test data set, and obtaining a test process emotion recognition result based on the test process video data set and a pre-trained micro-expression recognition model; Acquire a target time interval corresponding to the evaluation instruction, and acquire a current student user course data set and a current student user test score data set corresponding to the student user information and the target time interval; Based on the current student user course data set, the current student user test score data set and a pre-trained second classification model, a second current classification result is obtained; Based on the first current classification result, the test process emotion recognition result, the second current classification result and the pre-trained third classification model, a current classification output result corresponding to the student user information is obtained.

[0005] In a second aspect, an embodiment of the present invention provides a risk assessment device based on user psychological assessment data, which includes: A student user information acquisition unit, configured to acquire, in response to an evaluation instruction, student user information of a student object to be evaluated in the evaluation instruction; wherein the student user information at least includes a student user unique identification number, a student user tag, and a student historical evaluation result; A current test data set generating unit, configured to generate a current test data set based on a pre-built psychological assessment knowledge base, the student user information and a preset test data generating strategy, and send the current test data set to a receiving terminal corresponding to the student user information; A first classification unit, configured to receive a current feedback data set sent by the receiving terminal and corresponding to the current test data set, and obtain a first current classification result based on the current feedback data set and a pre-trained first classification model; An emotion recognition unit, configured to receive a test process video data set sent by the receiving terminal and corresponding to the current test data set, and obtain a test process emotion recognition result based on the test process video data set and a pre-trained micro-expression recognition model; A student course test data acquisition unit, used to acquire a target time interval corresponding to the evaluation instruction, and acquire a current student user course data set corresponding to the student user information and the target time interval, and a current student user test score data set; A second classification unit, used for obtaining a second current classification result based on the current student user course data set, the current student user test score data set and a pre-trained second classification model; The third classification unit is used to obtain the current classification output result corresponding to the student user information based on the first current classification result, the emotion recognition result of the test process, the second current classification result and the pre-trained third classification model.

[0006] In a third aspect, an embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer device executes the computer program, the risk assessment method based on user psychological assessment data as described in the first aspect above is implemented.

[0007] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the risk assessment method based on user psychological assessment data as described in the first aspect above is implemented.

[0008] The embodiment of the present invention provides a risk assessment method and device based on user psychological assessment data, the method comprising: in response to an assessment instruction, obtaining student user information of a student object to be assessed in the assessment instruction; wherein the student user information at least includes a student user unique identification number, a student user label and a student historical assessment result; based on a pre-constructed psychological assessment knowledge base, the student user information and a preset test data generation strategy, generating a current test data set, and sending the current test data set to a receiving terminal corresponding to the student user information; receiving a current feedback data set sent by the receiving terminal and corresponding to the current test data set, and obtaining a first current classification result based on the current feedback data set and a pre-trained first classification model; receiving a first current classification result received by a receiving terminal; Receive the test process video data set sent by the terminal and corresponding to the current test data set, and obtain the test process emotion recognition result based on the test process video data set and the pre-trained micro-expression recognition model; obtain the target time interval corresponding to the evaluation instruction, and obtain the current student user course data set corresponding to the student user information and the target time interval, as well as the current student user test score data set; obtain the second current classification result based on the current student user course data set, the current student user test score data set and the pre-trained second classification model; obtain the current classification output result corresponding to the student user information based on the first current classification result, the test process emotion recognition result, the second current classification result and the pre-trained third classification model. Through the above method, the current feedback data set of the student user's psychological assessment, the emotion recognition result corresponding to the test process video data set, the current student user course data set and the current student user test score data set can be obtained in time, and the multi-dimensional data of the student user can be effectively combined for classification and evaluation, so as to obtain the current classification output result with high real-time performance and more intuitive quantitative display, so as to provide the mental health professional teacher user with viewing and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0010] Figure 1 A schematic diagram of an application scenario of a risk assessment method based on user psychological assessment data provided by an embodiment of the present invention; Figure 2 A schematic diagram of a process for risk assessment based on user psychological assessment data provided by an embodiment of the present invention; Figure 3 A schematic diagram of a sub-process of a risk assessment method based on user psychological assessment data provided by an embodiment of the present invention; Figure 4 A schematic diagram of a sub-process of a risk assessment method based on user psychological assessment data provided by an embodiment of the present invention; Figure 5 A schematic diagram of a sub-process of a risk assessment method based on user psychological assessment data provided by an embodiment of the present invention; Figure 6 A schematic diagram of a sub-process of a risk assessment method based on user psychological assessment data provided by an embodiment of the present invention; Figure 7 A schematic diagram of a sub-process of a risk assessment method based on user psychological assessment data provided by an embodiment of the present invention; Figure 8 A schematic diagram of a sub-process of a risk assessment method based on user psychological assessment data provided by an embodiment of the present invention; Fig. 9 A schematic block diagram of a risk assessment device based on user psychological assessment data provided by an embodiment of the present invention; Fig.10 A schematic block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0011] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0012] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0013] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0014] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0015] See also Figure 1 and Figure 2 , Figure 1 The embodiment of the present invention provides Figure 1 A schematic diagram of an application scenario of a risk assessment method based on user psychological assessment data provided by an embodiment of the present invention; Figure 2 Schematic diagram of the process of the risk assessment method based on user psychological assessment data provided by the embodiment of the present invention; the risk assessment method based on user psychological assessment data is applied to the server 10, and the server 10 is connected to the first type of user terminal 20 used by the student user and the second type of user terminal 30 used by the teacher user. Figure 2 As shown, the method includes steps S110 to S170.

[0016] S110. In response to an evaluation instruction, obtain student user information of a student object to be evaluated in the evaluation instruction.

[0017] The student user information includes at least the student user's unique identification number, the student user tag and the student's historical evaluation results.

[0018] In this embodiment, when the teacher user needs to evaluate, collect and intelligently analyze the psychological state data of a student or multiple students in the recent period of time, the second type of user terminal can be used to establish a communication connection with the server. At this time, the teacher user can view the user interaction interface provided by the server on the second type of user terminal. The user interaction interface at least includes the logged-in teacher user information (such as teacher name, teacher number, teacher teaching subject, etc.), the teaching class corresponding to the teacher user information, and the student information list corresponding to each teaching class (including multiple student user information, and each student user corresponds to a student user). When the teacher user selects at least one student user information from the student information list including multiple student user information, and then clicks the evaluation virtual button located on the side of the student user information on the user interaction interface, the evaluation instruction corresponding to the student user information is triggered, and the evaluation instruction generation time corresponding to the evaluation instruction is also known. After the evaluation instruction is sent to the server by the second user terminal, the server can obtain the student user information of the student object to be evaluated in the evaluation instruction, which at least includes the user's unique identification number, the student user label and the student's historical evaluation results (corresponding to the evaluation results of the student user's previous participation in multiple assessments).

[0019] S120, generating a current test data set based on a pre-built psychological assessment knowledge base, the student user information and a preset test data generation strategy, and sending the current test data set to a receiving terminal corresponding to the student user information.

[0020] In this embodiment, after the server obtains the student user information of the student user to be evaluated, in order to realize the generation of the current test data set of non-fixed test questions for different student users, the student historical evaluation results, student user tags and other information in the student user information and the test data generation strategy can be combined to filter out data with multiple psychological assessment test questions from the psychological assessment knowledge base, thereby forming the current test data set. Moreover, after the current test data set is generated in the server, the communication information (such as mobile phone number, etc.) of the first type user terminal corresponding to the student user can be obtained according to the student user contact information corresponding to the student user information, and the first type user terminal is used as a receiving terminal to receive the current test data set sent by the server based on the communication information of the first type user terminal. For example, the server can send the web link address or push notification information corresponding to the current test data set to the receiving terminal, and the student user can quickly obtain the current test data set based on the receiving terminal, and view and feedback the test data for the current test data set.

[0021] In one embodiment, if Figure 3 As shown, step S120 includes: S121, obtaining a sub-strategy for determining the weight of the multi-dimensional to-be-assessed in the test data generation strategy, and generating a plurality of factors to be assessed and a factor weight value corresponding to each factor to be assessed based on the student user tag in the student user information, the historical risk assessment level corresponding to the student historical assessment result, and the sub-strategy for determining the weight of the multi-dimensional to-be-assessed; S122, obtaining a preset total number of test data, and determining the number of test data sub-items corresponding to each factor to be evaluated according to the product of the factor weight value corresponding to each factor to be evaluated among the multiple factors to be evaluated and the total number of test data; S123, for each of the multiple factors to be evaluated, according to the number of test data sub-items corresponding to the factor to be evaluated, randomly select a corresponding number of test data from the psychological assessment knowledge sub-base corresponding to the factor to be evaluated in the psychological assessment knowledge base, and form a current test data subset corresponding to the factor to be evaluated; S124: Fill the acquired current test data subset corresponding to each of the multiple factors to be evaluated into a preset test data template to obtain the current test data set.

[0022] In this embodiment, when the server generates the current test data set in combination with the student user tag, the historical risk assessment level and the test data generation strategy, the sub-strategy for determining the multi-dimensional weights to be evaluated in the test data generation strategy can be first obtained, and the sub-strategy for determining the multi-dimensional weights to be evaluated can be used to determine the multiple factors to be evaluated corresponding to the student user tag in the student user information and the historical risk assessment level of the student's historical assessment results, as well as the factor weight value corresponding to each factor to be evaluated.

[0023] Among them, the factor weight value of each factor to be evaluated determines the number of test data sub-items selected from the psychological assessment knowledge base later, and the factor weight value of the factor to be evaluated is greater than the number of test data sub-items corresponding to the factor to be evaluated. More specifically, the factor weight value corresponding to each factor to be evaluated in the multiple factors to be evaluated is multiplied by the total number of test data (such as setting the total number of test data to 25, 50, 100, 150, etc. according to actual user needs, of course, the above values ​​are only used as examples, and other custom values ​​can be used in specific implementations), so as to obtain the number of test data sub-items corresponding to each factor to be evaluated.

[0024] The psychological assessment knowledge base also includes multiple psychological assessment knowledge sub-bases, and each psychological assessment knowledge sub-base uniquely corresponds to a factor to be assessed. For example, the assessment factor base includes at least three comprehensive historical factors, disease factors, life stress event factors, somatization factors, personality characteristics factors, interpersonal factors, emotional situation factors, learning situation factors, cognitive coping factors and other factors, and the factor to be assessed is any factor in the assessment factor base. For example, after determining the number of test data sub-strips corresponding to a factor to be assessed (disease factor), the corresponding number of test data can be randomly screened from the psychological assessment knowledge sub-base corresponding to the factor to be assessed in the psychological assessment knowledge base according to the number of test data sub-strips, thereby forming a current test data subset corresponding to the factor to be assessed. The process of obtaining the current test data subsets corresponding to other factors to be assessed is also referred to the above process.

[0025] After finally obtaining the current test data subset corresponding to each of the multiple factors to be evaluated, a test data template can also be obtained, and the current test data subset corresponding to each factor to be evaluated is filled into the corresponding position in the test data template, thereby obtaining the current test data set that can be sent to student users for viewing and feedback.

[0026] In one embodiment, the sub-strategy for determining the weight of the multi-dimensional to be evaluated is a decision tree model, such as Figure 4 As shown, step S121 includes: S1211, obtaining current student user input data consisting of the student user label and the historical risk assessment level, and inputting the current student user data into the decision tree model corresponding to the multi-dimensional weight determination sub-strategy to be evaluated, to obtain a decision tree output result; S1212. Acquire multiple factors to be evaluated from a preset evaluation factor library according to the output result of the decision tree, and determine a factor weight value corresponding to each of the multiple factors to be evaluated based on the output result of the decision tree.

[0027] In this embodiment, the sub-strategy for determining the weight of the multi-dimensional to-be-assessed set in the server can specifically adopt a decision tree model, and the current student user input data consisting of the student user label and the historical risk assessment level (generally the first level to the third level, and the first level corresponds to the high risk level, the second level corresponds to the medium risk level, and the third level corresponds to the low risk level) is input into the decision tree model to obtain the decision tree output result. For example, the decision tree output result is a 1*9 vector such as [0.1, 0.2, 0.11, 0.13, 0.12, 0.05, 0.1, 0.09, 0.1], in which each non-zero value corresponds to a factor weight value of a factor to be assessed in the assessment factor library. For example, the second vector value 0.2 in the above 1*9 vector corresponds to the disease factor in the assessment factor library, and the factor weight value corresponding to the disease factor is 0.2. It can be seen that through the above method, multiple factors to be evaluated corresponding to student users and the factor weight value corresponding to each factor to be evaluated can be quickly determined based on the student user label and the historical risk assessment level, so as to be used for subsequent random screening from the psychological assessment knowledge sub-base corresponding to the factors to be evaluated to obtain a corresponding number of test data, thereby forming a current test data subset corresponding to the factors to be evaluated.

[0028] S130: Receive a current feedback data set sent by the receiving terminal and corresponding to the current test data set, and obtain a first current classification result based on the current feedback data set and a pre-trained first classification model.

[0029] In this embodiment, after the receiving terminal receives the current feedback data set sent by the server, the student user corresponding to the receiving terminal can use the receiving terminal, which is the first type of user terminal, to specifically view it, and can also perform a reply operation for each current test data in the current test data set. More specifically, the image acquisition device of the first type of user terminal, such as a camera, can also obtain the video data of the test process (including multiple user face images) when the student user performs a reply operation for each current test data in the current feedback data set. After the student user performs a reply operation for each current test data in the current feedback data set and obtains the corresponding current feedback data, the current feedback data set is formed and sent to the server.

[0030] Since the first classification model is pre-trained in the server (for example, a classification model such as a random forest model is specifically used), after the server receives the current feedback data set, a current first input vector can be formed by the identification value corresponding to each current feedback data in the current feedback data set, and input into the first classification model to obtain the first current classification result. It can be seen that through the above method, the first current classification result corresponding to the previous test data set can be quickly determined based on the first classification model.

[0031] In one embodiment, if Figure 5 As shown, step S130 includes: S131, obtaining a unique number and an identification value of a test data knowledge base corresponding to each current feedback data in the current feedback data set, and forming a current first input vector; S132: Input the current first input vector into the first classification model to obtain the first current classification result.

[0032] In this embodiment, each current test data in the current test data set corresponds to a current feedback data, and each current test data has a corresponding test data knowledge base unique number in the psychological assessment knowledge base, and different current feedback data for each current test data corresponds to different identification values. When the current feedback data after the student user responds to each current test data in the current test data set is obtained, the test data knowledge base unique number and identification value corresponding to each current feedback data in the current feedback data set can be quickly obtained, and the current first input vector is formed. After that, the current first input vector is input into the first classification model, which is a random forest model, and the first current classification result is obtained. For example, in the specific implementation, the first current classification result is a first probability value with a value range of 0-1. When the value of the first current classification result is 0~1 / 3, it corresponds to a low risk assessment result of the psychological assessment, when the value of the first current classification result is 1 / 3~2 / 3, it corresponds to a medium risk assessment result of the psychological assessment, and when the value of the first current classification result is 2 / 3~1, it corresponds to a high risk assessment result of the psychological assessment.

[0033] S140, receiving a test process video data set sent by the receiving terminal and corresponding to the current test data set, and obtaining a test process emotion recognition result based on the test process video data set and a pre-trained micro-expression recognition model.

[0034] In this embodiment, an image acquisition device such as a camera based on the receiving terminal, which is a first type of user terminal, obtains the video data of the test process when the student user performs a reply operation on each current test data, and a test process video data set can be formed. Moreover, based on the micro-expression recognition model, the micro-expression recognition result corresponding to each current feedback data can be obtained by performing micro-expression recognition on the test process video data corresponding to each current test data in the server. Afterwards, the micro-expression recognition result corresponding to each current feedback data is comprehensively combined to perform statistics on the highest frequency micro-expression recognition result to obtain the test process emotion recognition result.

[0035] In one embodiment, if Figure 6 As shown, step S140 includes: S141, for each test process video data corresponding to the current test data in the current test data set, obtaining the corresponding target user facial image based on a preset key frame image extraction strategy, and performing micro-expression recognition on the target user facial image through the micro-expression recognition model to obtain a micro-expression recognition result; S142, obtaining the micro-expression recognition result with the maximum number of recognition result statistics among the micro-expression recognition results corresponding to each current test data as the emotion recognition result of the test process.

[0036] In this embodiment, for each current test data in the current test data set, the test process video data corresponding to each current feedback data in the current feedback data set is obtained. At this time, the micro-expression recognition result is obtained for the test process video data corresponding to a current feedback data as an example. The test process video data corresponding to the current feedback data is essentially a multi-frame user image. At this time, one of the target user facial images including the user's face image can be selected based on the key frame image extraction strategy, and then the target user facial image is subjected to micro-expression recognition through the micro-expression recognition model (such as the Micron-BERT facial micro-expression recognition model, etc.) to obtain the micro-expression recognition result. After obtaining the micro-expression recognition results corresponding to each current test data in the current test data set, the micro-expression recognition result with the highest frequency is obtained and used as the test process emotion recognition result. It can be seen that through the above method, the micro-expression recognition of the current feedback data set of the student user in the process of feedback reply to the current test data set can be realized, and the emotion recognition result of the user's reply process can be obtained.

[0037] S150, obtaining a target time interval corresponding to the evaluation instruction, and obtaining a current student user course data set and a current student user test score data set corresponding to the student user information and the target time interval.

[0038] In this embodiment, when the server receives the evaluation instruction, it can also obtain the evaluation instruction generation time corresponding to the evaluation instruction. The server also presets a preset evaluation cycle duration (such as one week, half a month, one month, three months, half a year, one year, etc.). At this time, the evaluation instruction generation time-preset evaluation cycle duration is used as the starting time point of the target time interval, and the evaluation instruction generation time is used as the end time point of the target time interval, thereby obtaining the target time interval. Afterwards, the current student user course data set corresponding to the student user information and the target time interval, as well as the current student user test score data set, are obtained in the local database of the server as a reference data set for subsequent analysis of the student user's recent psychological evaluation status.

[0039] S160. Obtain a second current classification result based on the current student user course data set, the current student user test score data set and a pre-trained second classification model.

[0040] In this embodiment, after obtaining the current student user course data set and the current student user test score data set, it is possible to obtain which specific courses the student user has recently studied, as well as specific test score data for different examination subjects. The process of obtaining the first current classification result based on the current feedback data set and the pre-trained first classification model can be referred to above, which is also to form a current second input vector by combining the current student user course data set and the current student user test score data set. Afterwards, the current second input vector is input into the second classification model to obtain the second current classification result.

[0041] In one embodiment, if Figure 7 As shown, step S160 includes: S161, obtaining the course unique number corresponding to each current student user course data in the current student user course data set, and the student user test score data corresponding to each course unique number, and forming the current second input vector; S162. Input the current second input vector into the second classification model to obtain the second current classification result.

[0042] In this embodiment, the unique course number corresponding to each current student user course data in the current student user course data set, and the student user test score data corresponding to each current student user course data can be obtained in the current student user test score data set, and form the current second input vector. Afterwards, the current second input vector is input into the second classification model, which is a random forest model, and the second current classification result is obtained. For example, in the specific implementation, the second current classification result is a second probability value in the range of 0-1. When the value of the second current classification result is 0~1 / 3, it corresponds to the low risk assessment result of course learning and test pressure. When the value of the second current classification result is 1 / 3~2 / 3, it corresponds to the medium risk assessment result of course learning and test pressure. When the value of the second current classification result is 2 / 3~1, it corresponds to the high risk assessment result of course learning and test pressure.

[0043] Among them, it should be noted that the first current classification result is the risk assessment classification result determined from the psychological assessment dimension, while the second current classification result is the risk assessment classification result determined from the course learning and test pressure dimension.

[0044] S170. Based on the first current classification result, the test process emotion recognition result, the second current classification result and the pre-trained third classification model, obtain the current classification output result corresponding to the student user information.

[0045] In this embodiment, when the risk assessment classification results are determined from the psychological assessment dimension and the course learning and test pressure dimension respectively, they can also be combined with the emotion recognition results of the test process during the psychological assessment test. The three together form a current third input vector and input it into the third classification model to obtain the current classification output result corresponding to the student user information.

[0046] In one embodiment, if Figure 8 As shown, step S170 includes: S171, obtaining a first probability value corresponding to the first current classification result, a result identification value corresponding to the emotion recognition result of the test process, and a second probability value corresponding to the second current classification result, and forming a current third input vector; S172, inputting the current third input vector into the third classification model to obtain a third current classification result; S173. Based on the third current classification result and the pre-constructed mapping relationship between classification results and evaluation results, determine a target evaluation result and use it as the current classification output result.

[0047] In this embodiment, after obtaining the first probability value corresponding to the first current classification result, the result identification value corresponding to the emotion recognition result of the test process, and the second probability value corresponding to the second current classification result, the current third input vector is formed in the above order, and input into the third classification model, which is a random forest model, to obtain the third current classification result. For example, in the specific implementation, a plurality of third current classification results and the evaluation results corresponding to each of the third current classification results are pre-set, and a mapping relationship between the classification results and the evaluation results is formed. After the third current classification result is currently determined, the target evaluation result can be obtained by matching and used as the current classification output result. More specifically, the mapping relationship between the classification result and the evaluation result includes two third current classification results, which are the evaluation results that require external intervention and adjustment and the evaluation results that require slight adjustment by themselves. Moreover, the obtained current classification output result can be sent by the server to the second type of user terminal corresponding to the teacher user for viewing in a timely manner.

[0048] It can be seen that the implementation example of this method can timely obtain the current feedback data set of student users' psychological assessment, the emotion recognition results corresponding to the test process video data set, the current student user course data set, and the current student user test score data set, and effectively combine the multi-dimensional data of student users for classification and evaluation, and obtain the current classification output results with high real-time performance and more intuitive quantitative display for viewing and analysis by mental health professional teacher users.

[0049] The embodiment of the present invention also provides a risk assessment device based on user psychological assessment data, which can be configured in a server and is used to execute any embodiment of the risk assessment method based on user psychological assessment data. Fig. 9 , Fig. 9 Schematic block diagram of a risk assessment device based on user psychological assessment data provided by an embodiment of the present invention. Fig. 9 As shown, the risk assessment device 100 based on user psychological assessment data includes a student user information acquisition unit 110, a current test data set generation unit 120, a first classification unit 130, an emotion recognition unit 140, a student course test data acquisition unit 150, a second classification unit 160 and a third classification unit 170.

[0050] The student user information acquisition unit 110 is used to acquire the student user information of the student object to be evaluated in response to the evaluation instruction.

[0051] The student user information includes at least the student user's unique identification number, the student user tag and the student's historical evaluation results.

[0052] In this embodiment, when the teacher user needs to evaluate, collect and intelligently analyze the psychological state data of a student or multiple students in the recent period of time, the second type of user terminal can be used to establish a communication connection with the server. At this time, the teacher user can view the user interaction interface provided by the server on the second type of user terminal. The user interaction interface at least includes the logged-in teacher user information (such as teacher name, teacher number, teacher teaching subject, etc.), the teaching class corresponding to the teacher user information, and the student information list corresponding to each teaching class (including multiple student user information, and each student user corresponds to a student user). When the teacher user selects at least one student user information from the student information list including multiple student user information, and then clicks the evaluation virtual button located on the side of the student user information on the user interaction interface, the evaluation instruction corresponding to the student user information is triggered, and the evaluation instruction generation time corresponding to the evaluation instruction is also known. After the evaluation instruction is sent to the server by the second user terminal, the server can obtain the student user information of the student object to be evaluated in the evaluation instruction, which at least includes the user's unique identification number, the student user label and the student's historical evaluation results (corresponding to the evaluation results of the student user's previous participation in multiple assessments).

[0053] The current test data set generating unit 120 is used to generate a current test data set based on a pre-built psychological assessment knowledge base, the student user information and a preset test data generating strategy, and send the current test data set to a receiving terminal corresponding to the student user information.

[0054] In this embodiment, after the server obtains the student user information of the student user to be evaluated, in order to realize the generation of the current test data set of non-fixed test questions for different student users, the student historical evaluation results, student user tags and other information in the student user information and the test data generation strategy can be combined to filter out data with multiple psychological assessment test questions from the psychological assessment knowledge base, thereby forming the current test data set. Moreover, after the current test data set is generated in the server, the communication information (such as mobile phone number, etc.) of the first type user terminal corresponding to the student user can be obtained according to the student user contact information corresponding to the student user information, and the first type user terminal is used as a receiving terminal to receive the current test data set sent by the server based on the communication information of the first type user terminal. For example, the server can send the web link address or push notification information corresponding to the current test data set to the receiving terminal, and the student user can quickly obtain the current test data set based on the receiving terminal, and view and feedback the test data for the current test data set.

[0055] In one embodiment, the current test data set generating unit 120 is specifically used for: Obtain the sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated in the test data generation strategy, and generate multiple factors to be evaluated and factor weight values ​​corresponding to each factor to be evaluated based on the student user label in the student user information, the historical risk assessment level corresponding to the student historical assessment result, and the sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated; Obtaining a preset total number of test data, and determining the number of test data sub-items corresponding to each factor to be evaluated according to the product of the factor weight value corresponding to each factor to be evaluated among the multiple factors to be evaluated and the total number of test data; For each of the multiple factors to be evaluated, according to the number of test data sub-items corresponding to the factor to be evaluated, randomly select a corresponding number of test data from the psychological assessment knowledge sub-base corresponding to the factor to be evaluated in the psychological assessment knowledge base to form a current test data subset corresponding to the factor to be evaluated; The acquired current test data subset corresponding to each of the multiple factors to be evaluated is filled into a preset test data template to obtain the current test data set.

[0056] In this embodiment, when the server generates the current test data set in combination with the student user tag, the historical risk assessment level and the test data generation strategy, the sub-strategy for determining the multi-dimensional weights to be evaluated in the test data generation strategy can be first obtained, and the sub-strategy for determining the multi-dimensional weights to be evaluated can be used to determine the multiple factors to be evaluated corresponding to the student user tag in the student user information and the historical risk assessment level of the student's historical assessment results, as well as the factor weight value corresponding to each factor to be evaluated.

[0057] Among them, the factor weight value of each factor to be evaluated determines the number of test data sub-items selected from the psychological assessment knowledge base later, and the factor weight value of the factor to be evaluated is greater than the number of test data sub-items corresponding to the factor to be evaluated. More specifically, the factor weight value corresponding to each factor to be evaluated in the multiple factors to be evaluated is multiplied by the total number of test data (such as setting the total number of test data to 25, 50, 100, 150, etc. according to actual user needs, of course, the above values ​​are only used as examples, and other custom values ​​can be used in specific implementations), so as to obtain the number of test data sub-items corresponding to each factor to be evaluated.

[0058] The psychological assessment knowledge base also includes multiple psychological assessment knowledge sub-bases, and each psychological assessment knowledge sub-base uniquely corresponds to a factor to be assessed. For example, the assessment factor base includes at least three comprehensive historical factors, disease factors, life stress event factors, somatization factors, personality characteristics factors, interpersonal factors, emotional situation factors, learning situation factors, cognitive coping factors and other factors, and the factor to be assessed is any factor in the assessment factor base. For example, after determining the number of test data sub-strips corresponding to a factor to be assessed (disease factor), the corresponding number of test data can be randomly screened from the psychological assessment knowledge sub-base corresponding to the factor to be assessed in the psychological assessment knowledge base according to the number of test data sub-strips, thereby forming a current test data subset corresponding to the factor to be assessed. The process of obtaining the current test data subsets corresponding to other factors to be assessed is also referred to the above process.

[0059] After finally obtaining the current test data subset corresponding to each of the multiple factors to be evaluated, a test data template can also be obtained, and the current test data subset corresponding to each factor to be evaluated is filled into the corresponding position in the test data template, thereby obtaining the current test data set that can be sent to student users for viewing and feedback.

[0060] In one embodiment, the sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated is a decision tree model, and the sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated based on the student user label in the student user information, the historical risk assessment level corresponding to the student historical assessment result, and the sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated generates multiple factors to be evaluated and factor weight values ​​corresponding to each factor to be evaluated, including: Obtaining current student user input data consisting of the student user label and the historical risk assessment level, and inputting the current student user data into the decision tree model corresponding to the multi-dimensional weight determination sub-strategy to be evaluated, to obtain a decision tree output result; According to the output result of the decision tree, a plurality of factors to be evaluated are obtained from a preset evaluation factor library, and based on the output result of the decision tree, a factor weight value corresponding to each of the plurality of factors to be evaluated is determined.

[0061] In this embodiment, the sub-strategy for determining the weight of the multi-dimensional to-be-assessed set in the server can specifically adopt a decision tree model, and the current student user input data consisting of the student user label and the historical risk assessment level (generally the first level to the third level, and the first level corresponds to the high risk level, the second level corresponds to the medium risk level, and the third level corresponds to the low risk level) is input into the decision tree model to obtain the decision tree output result. For example, the decision tree output result is a 1*9 vector such as [0.1, 0.2, 0.11, 0.13, 0.12, 0.05, 0.1, 0.09, 0.1], in which each non-zero value corresponds to a factor weight value of a factor to be assessed in the assessment factor library. For example, the second vector value 0.2 in the above 1*9 vector corresponds to the disease factor in the assessment factor library, and the factor weight value corresponding to the disease factor is 0.2. It can be seen that through the above method, multiple factors to be evaluated corresponding to student users and the factor weight value corresponding to each factor to be evaluated can be quickly determined based on the student user label and the historical risk assessment level, so as to be used for subsequent random screening from the psychological assessment knowledge sub-base corresponding to the factors to be evaluated to obtain a corresponding number of test data, thereby forming a current test data subset corresponding to the factors to be evaluated.

[0062] The first classification unit 130 is configured to receive a current feedback data set sent by the receiving terminal and corresponding to the current test data set, and obtain a first current classification result based on the current feedback data set and a pre-trained first classification model.

[0063] In this embodiment, after the receiving terminal receives the current feedback data set sent by the server, the student user corresponding to the receiving terminal can use the receiving terminal, which is the first type of user terminal, to specifically view it, and can also perform a reply operation for each current test data in the current test data set. More specifically, the image acquisition device of the first type of user terminal, such as a camera, can also obtain the video data of the test process (including multiple user face images) when the student user performs a reply operation for each current test data in the current feedback data set. After the student user performs a reply operation for each current test data in the current feedback data set and obtains the corresponding current feedback data, the current feedback data set is formed and sent to the server.

[0064] Since the first classification model is pre-trained in the server (for example, a classification model such as a random forest model is specifically used), after the server receives the current feedback data set, a current first input vector can be formed by the identification value corresponding to each current feedback data in the current feedback data set, and input into the first classification model to obtain the first current classification result. It can be seen that through the above method, the first current classification result corresponding to the previous test data set can be quickly determined based on the first classification model.

[0065] In one embodiment, the first classification unit 130 is specifically used for: Acquire a test data knowledge base unique number and an identification value corresponding to each current feedback data in the current feedback data set, and form a current first input vector; The current first input vector is input into the first classification model to obtain the first current classification result.

[0066] In this embodiment, each current test data in the current test data set corresponds to a current feedback data, and each current test data has a corresponding test data knowledge base unique number in the psychological assessment knowledge base, and different current feedback data for each current test data corresponds to different identification values. When the current feedback data after the student user responds to each current test data in the current test data set is obtained, the test data knowledge base unique number and identification value corresponding to each current feedback data in the current feedback data set can be quickly obtained, and the current first input vector is formed. After that, the current first input vector is input into the first classification model, which is a random forest model, and the first current classification result is obtained. For example, in the specific implementation, the first current classification result is a first probability value with a value range of 0-1. When the value of the first current classification result is 0~1 / 3, it corresponds to a low risk assessment result of the psychological assessment, when the value of the first current classification result is 1 / 3~2 / 3, it corresponds to a medium risk assessment result of the psychological assessment, and when the value of the first current classification result is 2 / 3~1, it corresponds to a high risk assessment result of the psychological assessment.

[0067] The emotion recognition unit 140 is used to receive a test process video data set sent by the receiving terminal and corresponding to the current test data set, and obtain a test process emotion recognition result based on the test process video data set and a pre-trained micro-expression recognition model.

[0068] In this embodiment, an image acquisition device such as a camera based on the receiving terminal, which is a first type of user terminal, obtains the video data of the test process when the student user performs a reply operation on each current test data, and a test process video data set can be formed. Moreover, based on the micro-expression recognition model, the micro-expression recognition result corresponding to each current feedback data can be obtained by performing micro-expression recognition on the test process video data corresponding to each current test data in the server. Afterwards, the micro-expression recognition result corresponding to each current feedback data is comprehensively combined to perform statistics on the highest frequency micro-expression recognition result to obtain the test process emotion recognition result.

[0069] In one embodiment, the emotion recognition unit 140 is specifically used for: For each test process video data corresponding to the current test data in the current test data set, the corresponding target user facial image is obtained based on a preset key frame image extraction strategy, and micro-expression recognition is performed on the target user facial image through the micro-expression recognition model to obtain a micro-expression recognition result; The micro-expression recognition result with the maximum number of recognition result statistics among the micro-expression recognition results corresponding to each current test data is obtained as the emotion recognition result of the test process.

[0070] In this embodiment, for each current test data in the current test data set, the test process video data corresponding to each current feedback data in the current feedback data set is obtained. At this time, the micro-expression recognition result is obtained for the test process video data corresponding to a current feedback data as an example. The test process video data corresponding to the current feedback data is essentially a multi-frame user image. At this time, one of the target user facial images including the user's face image can be selected based on the key frame image extraction strategy, and then the target user facial image is subjected to micro-expression recognition through the micro-expression recognition model (such as Micron-BERT facial micro-expression recognition model, etc.) to obtain the micro-expression recognition result. After obtaining the micro-expression recognition results corresponding to each current test data in the current test data set, the micro-expression recognition result with the highest frequency is obtained and used as the test process emotion recognition result. It can be seen that through the above method, the micro-expression recognition of the current feedback data set of the student user in the process of feedback reply to the current test data set can be realized, and the emotion recognition result of the user's reply process can be obtained.

[0071] The student course test data acquisition unit 150 is used to acquire the target time interval corresponding to the evaluation instruction, and acquire the current student user course data set corresponding to the student user information and the target time interval, and the current student user test score data set.

[0072] In this embodiment, when the server receives the evaluation instruction, it can also obtain the evaluation instruction generation time corresponding to the evaluation instruction. The server also presets a preset evaluation cycle duration (such as one week, half a month, one month, three months, half a year, one year, etc.). At this time, the evaluation instruction generation time-preset evaluation cycle duration is used as the starting time point of the target time interval, and the evaluation instruction generation time is used as the end time point of the target time interval, thereby obtaining the target time interval. Afterwards, the current student user course data set corresponding to the student user information and the target time interval, as well as the current student user test score data set, are obtained in the local database of the server as a reference data set for subsequent analysis of the student user's recent psychological evaluation status.

[0073] The second classification unit 160 is used to obtain a second current classification result based on the current student user course data set, the current student user test score data set and a pre-trained second classification model.

[0074] In this embodiment, after obtaining the current student user course data set and the current student user test score data set, it is possible to obtain which specific courses the student user has recently studied, as well as specific test score data for different examination subjects. The process of obtaining the first current classification result based on the current feedback data set and the pre-trained first classification model can be referred to above, which is also to form a current second input vector by combining the current student user course data set and the current student user test score data set. Afterwards, the current second input vector is input into the second classification model to obtain the second current classification result.

[0075] In one embodiment, the second classification unit 160 is specifically used for: Obtain the course unique number corresponding to each current student user course data in the current student user course data set, and the student user test score data corresponding to each course unique number, and form the current second input vector; The current second input vector is input into the second classification model to obtain the second current classification result.

[0076] In this embodiment, the unique course number corresponding to each current student user course data in the current student user course data set, and the student user test score data corresponding to each current student user course data can be obtained in the current student user test score data set, and form the current second input vector. Afterwards, the current second input vector is input into the second classification model, which is a random forest model, and the second current classification result is obtained. For example, in the specific implementation, the second current classification result is a second probability value in the range of 0-1. When the value of the second current classification result is 0~1 / 3, it corresponds to the low risk assessment result of course learning and test pressure. When the value of the second current classification result is 1 / 3~2 / 3, it corresponds to the medium risk assessment result of course learning and test pressure. When the value of the second current classification result is 2 / 3~1, it corresponds to the high risk assessment result of course learning and test pressure.

[0077] Among them, it should be noted that the first current classification result is the risk assessment classification result determined from the psychological assessment dimension, while the second current classification result is the risk assessment classification result determined from the course learning and test pressure dimension.

[0078] The third classification unit 170 is used to obtain the current classification output result corresponding to the student user information based on the first current classification result, the test process emotion recognition result, the second current classification result and the pre-trained third classification model.

[0079] In this embodiment, when the risk assessment classification results are determined from the psychological assessment dimension and the course learning and test pressure dimension respectively, they can also be combined with the emotion recognition results of the test process during the psychological assessment test. The three together form a current third input vector and input it into the third classification model to obtain the current classification output result corresponding to the student user information.

[0080] In one embodiment, the third classification unit 170 is specifically used for: Obtaining a first probability value corresponding to the first current classification result, a result identification value corresponding to the test process emotion recognition result, and a second probability value corresponding to the second current classification result, and forming a current third input vector; Inputting the current third input vector into the third classification model to obtain a third current classification result; Based on the third current classification result and the pre-constructed mapping relationship between the classification result and the evaluation result, a target evaluation result is determined and used as the current classification output result.

[0081] In this embodiment, after obtaining the first probability value corresponding to the first current classification result, the result identification value corresponding to the emotion recognition result of the test process, and the second probability value corresponding to the second current classification result, the current third input vector is formed in the above order, and input into the third classification model, which is a random forest model, to obtain the third current classification result. For example, in the specific implementation, a plurality of third current classification results and the evaluation results corresponding to each of the third current classification results are pre-set, and a mapping relationship between the classification results and the evaluation results is formed. After the third current classification result is currently determined, the target evaluation result can be obtained by matching and used as the current classification output result. More specifically, the mapping relationship between the classification result and the evaluation result includes two third current classification results, which are the evaluation results that require external intervention and adjustment and the evaluation results that require slight adjustment by themselves. Moreover, the obtained current classification output result can be sent by the server to the second type of user terminal corresponding to the teacher user for viewing in a timely manner.

[0082] It can be seen that the implementation example of the device can timely obtain the current feedback data set of student users' psychological assessment, the emotion recognition results corresponding to the test process video data set, the current student user course data set and the current student user test score data set, and effectively combine the multi-dimensional data of student users for classification and evaluation, and obtain the current classification output results with high real-time performance and more intuitive quantitative display for viewing and analysis by mental health professional teacher users.

[0083] The risk assessment device based on user psychological assessment data can be implemented in the form of a computer program. Fig.10 Runs on the computer device shown.

[0084] See also Fig.10 , Fig.10 1 is a schematic block diagram of a computer device provided by an embodiment of the present invention. The computer device integrates any risk assessment device based on user psychological assessment data provided by an embodiment of the present invention.

[0085] See also Fig.10 The computer device 400 includes a processor 402 , a memory and a network interface 405 connected via a system bus 401 , wherein the memory may include a storage medium 403 and an internal memory 404 .

[0086] The storage medium 403 can store an operating system 4031 and a computer program 4032. The computer program 4032 includes program instructions, and when the program instructions are executed, the processor 402 can execute the above-mentioned risk assessment method based on user psychological assessment data.

[0087] The processor 402 is used to provide computing and control capabilities to support the operation of the entire computer device.

[0088] The internal memory 404 provides an environment for the operation of the computer program 4032 in the storage medium 403. When the computer program 4032 is executed by the processor 402, the processor 402 can execute the above-mentioned risk assessment method based on the user psychological assessment data.

[0089] The network interface 405 is used to communicate with other devices over the network. Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0090] The processor 402 is used to run the computer program 4032 stored in the memory to implement the risk assessment method based on user psychological assessment data as described above.

[0091] It should be understood that in the embodiment of the present invention, the processor 402 may be a central processing unit (CPU), and the processor 402 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0092] It can be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment can be completed by instructing the relevant hardware through a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer device to implement the process steps of the embodiment of the above method.

[0093] Therefore, the present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions. When the program instructions are executed by a processor, the processor executes the risk assessment method based on user psychological assessment data as described above.

[0094] The computer-readable storage medium may be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk, etc., which are computer-readable storage media that can store program codes.

[0095] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0096] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of each unit is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0097] The steps in the method of the embodiment of the present invention can be adjusted in order, combined and deleted according to actual needs. The units in the device of the embodiment of the present invention can be combined, divided and deleted according to actual needs. In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0098] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention.

[0099] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A risk assessment method based on user psychological assessment data, characterized in that: include: In response to the evaluation instruction, obtaining the student user information of the student object to be evaluated in the evaluation instruction; wherein the student user information at least includes the student user unique identification number, the student user tag and the student historical evaluation result; Based on a pre-built psychological assessment knowledge base, the student user information and a preset test data generation strategy, a current test data set is generated, and the current test data set is sent to a receiving terminal corresponding to the student user information; Receiving a current feedback data set sent by the receiving terminal and corresponding to the current test data set, and acquiring a first current classification result based on the current feedback data set and a pre-trained first classification model; Receiving a test process video data set sent by the receiving terminal and corresponding to the current test data set, and obtaining a test process emotion recognition result based on the test process video data set and a pre-trained micro-expression recognition model; Acquire a target time interval corresponding to the evaluation instruction, and acquire a current student user course data set and a current student user test score data set corresponding to the student user information and the target time interval; Based on the current student user course data set, the current student user test score data set and a pre-trained second classification model, a second current classification result is obtained; Based on the first current classification result, the test process emotion recognition result, the second current classification result and the pre-trained third classification model, a current classification output result corresponding to the student user information is obtained.

2. The risk assessment method based on user psychological evaluation data according to claim 1 is characterized in that: The generating of the current test data set based on the pre-built psychological assessment knowledge base, the student user information and the preset test data generating strategy includes: Obtain the sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated in the test data generation strategy, and generate multiple factors to be evaluated and factor weight values ​​corresponding to each factor to be evaluated based on the student user label in the student user information, the historical risk assessment level corresponding to the student historical assessment result, and the sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated; Obtaining a preset total number of test data, and determining the number of test data sub-items corresponding to each factor to be evaluated according to the product of the factor weight value corresponding to each factor to be evaluated among the multiple factors to be evaluated and the total number of test data; For each of the multiple factors to be evaluated, according to the number of test data sub-items corresponding to the factor to be evaluated, randomly select a corresponding number of test data from the psychological assessment knowledge sub-base corresponding to the factor to be evaluated in the psychological assessment knowledge base to form a current test data subset corresponding to the factor to be evaluated; The acquired current test data subset corresponding to each of the multiple factors to be evaluated is filled into a preset test data template to obtain the current test data set.

3. The risk assessment method based on user psychological evaluation data according to claim 2 is characterized in that: The sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated is a decision tree model; the sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated based on the student user label in the student user information, the historical risk assessment level corresponding to the student historical assessment result and the sub-strategy for determining the weight of the multi-dimensional dimensions to be evaluated generates multiple factors to be evaluated and factor weight values ​​corresponding to each factor to be evaluated, including: Obtaining current student user input data consisting of the student user label and the historical risk assessment level, and inputting the current student user data into the decision tree model corresponding to the multi-dimensional weight determination sub-strategy to be evaluated, to obtain a decision tree output result; According to the output result of the decision tree, a plurality of factors to be evaluated are obtained from a preset evaluation factor library, and based on the output result of the decision tree, a factor weight value corresponding to each of the plurality of factors to be evaluated is determined.

4. The risk assessment method based on user psychological evaluation data according to claim 1, characterized in that: The obtaining a first current classification result based on the current feedback data set and a pre-trained first classification model includes: Acquire a test data knowledge base unique number and an identification value corresponding to each current feedback data in the current feedback data set, and form a current first input vector; The current first input vector is input into the first classification model to obtain the first current classification result.

5. The risk assessment method based on user psychological evaluation data according to claim 1, characterized in that: The step of obtaining the emotion recognition result of the test process based on the test process video data set and the pre-trained micro-expression recognition model includes: For each test process video data corresponding to the current test data in the current test data set, the corresponding target user facial image is obtained based on a preset key frame image extraction strategy, and micro-expression recognition is performed on the target user facial image through the micro-expression recognition model to obtain a micro-expression recognition result; The micro-expression recognition result with the maximum number of recognition result statistics among the micro-expression recognition results corresponding to each current test data is obtained as the emotion recognition result of the test process.

6. The risk assessment method based on user psychological evaluation data according to claim 1, characterized in that: The obtaining of a second current classification result based on the current student user course data set, the current student user test score data set and the pre-trained second classification model includes: Obtain the course unique number corresponding to each current student user course data in the current student user course data set, and the student user test score data corresponding to each course unique number, and form the current second input vector; The current second input vector is input into the second classification model to obtain the second current classification result.

7. The risk assessment method based on user psychological evaluation data according to claim 1, characterized in that: The obtaining of a current classification output result corresponding to the student user information based on the first current classification result, the test process emotion recognition result, the second current classification result and a pre-trained third classification model includes: Obtaining a first probability value corresponding to the first current classification result, a result identification value corresponding to the emotion recognition result of the test process, and a second probability value corresponding to the second current classification result, and forming a current third input vector; Inputting the current third input vector into the third classification model to obtain a third current classification result; Based on the third current classification result and the pre-constructed mapping relationship between the classification result and the evaluation result, a target evaluation result is determined and used as the current classification output result.

8. A risk assessment device based on user psychological assessment data, characterized in that: include: A student user information acquisition unit, configured to acquire, in response to an evaluation instruction, student user information of a student object to be evaluated in the evaluation instruction; wherein the student user information at least includes a student user unique identification number, a student user tag, and a student historical evaluation result; A current test data set generating unit, configured to generate a current test data set based on a pre-built psychological assessment knowledge base, the student user information and a preset test data generating strategy, and send the current test data set to a receiving terminal corresponding to the student user information; A first classification unit, configured to receive a current feedback data set sent by the receiving terminal and corresponding to the current test data set, and obtain a first current classification result based on the current feedback data set and a pre-trained first classification model; An emotion recognition unit, configured to receive a test process video data set sent by the receiving terminal and corresponding to the current test data set, and obtain a test process emotion recognition result based on the test process video data set and a pre-trained micro-expression recognition model; A student course test data acquisition unit, used to acquire a target time interval corresponding to the evaluation instruction, and acquire a current student user course data set corresponding to the student user information and the target time interval, and a current student user test score data set; A second classification unit, used for obtaining a second current classification result based on the current student user course data set, the current student user test score data set and a pre-trained second classification model; The third classification unit is used to obtain the current classification output result corresponding to the student user information based on the first current classification result, the emotion recognition result of the test process, the second current classification result and the pre-trained third classification model.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the computer device executes the computer program, the risk assessment method based on user psychological assessment data as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the risk assessment method based on user psychological assessment data as described in any one of claims 1 to 7 is implemented.

Images