Method, system and electronic device for obtaining character psychological portrait based on questionnaire evaluation

Abstract

The present application relates to a method, system, and electronic device for obtaining a psychological portrait of a person based on a questionnaire assessment. The method includes: collecting multiple dimensional data of a psychological questionnaire assessment process; performing deep learning based on the multiple dimensional data of the assessment process, and collaborative clustering to obtain a psychological portrait of the assessor. Compared with simply conducting a questionnaire assessment, the psychological portrait of a person obtained by the present application is more accurate. At the same time, the above technical solution of the present application has broad social benefits and can fill the gap in the field of psychological assessment that only focuses on results but lacks process data.

Description

Technical Field

[0001] The present application relates to the technical field of character psychological portraits, and in particular to a method, system, and electronic device for obtaining character psychological portraits based on questionnaire evaluation. Background Art

[0002] As people's material living standards continue to improve, rapid development has also brought increasing pressure. In recent years, various domestic and international turmoil and trade challenges have continued to intensify, causing people to have psychological pressure beyond the past. The World Health Organization pointed out that the global burden brought by psychological problems will be the second most important cause of overall human morbidity. The National Health and Family Planning Commission and 22 other ministries and commissions jointly issued the "Guiding Opinions on Strengthening Mental Health Services", which mentioned the importance of paying attention to the psychological service system. The 13th Five-Year Plan Suggestions, the National Social Psychological Service System Construction Pilot Work Plan and the 19th National Congress Report mentioned strengthening the construction of the psychological service system and cultivating a positive social mentality.

[0003] Although the state and ministries have made top-level design and policy guidance for psychological services, there are still many deficiencies in the existing psychological services in China. The policy implementation is not perfect, and the design and implementation of psychological service-related projects (papers, patents or related topics) have not played their due role.

[0004] Although different manufacturers have developed corresponding assessment tools, which are generally divided into symptom diagnosis, personality measurement, and mental health assessment, due to different theoretical orientations, most of these software or apps provide homogeneous functions, and only the questionnaire results and simple arithmetic average results are output to users.

[0005] The lack of evaluation methods is one of the reasons for the unsatisfactory results of psychological assessment. It is particularly worth mentioning that the same group of students take the same type of psychological assessment, and the psychological assessment results of different manufacturers are far apart due to the different tools used. This can't help but make people wonder: Can the assessment results truly reflect the psychological status of the assessors, how to evaluate the quality of psychological assessment results, etc., so more efficient psychological assessment projects or research topics are needed. Professor Yang Zongkai proposed that information technology should be deeply integrated with education to achieve cross-integration results. Professor Chen Li of Beijing Normal University proposed that the new concepts, new technologies and new methods of Internet + education can be used to solve problems related to mental health services. At present, most of the existing assessments only record the results, but ignore the importance of process data in the assessment. For example, the existing well-known Questionnaire Star only provides users with simple statistics, without considering the cross-integration of information technology or the new methods of Internet +, so that process data that is helpful for result judgment cannot be obtained, and the final psychological portrait of the person obtained is not very accurate. Summary of the invention

[0006] In order to obtain a more accurate character psychological portrait based on questionnaire evaluation, the present application provides a character psychological portrait acquisition method, system, and electronic device based on questionnaire evaluation.

[0007] In the first aspect, the present application provides a method for obtaining a character psychological portrait based on a questionnaire assessment, which adopts the following technical solution:

[0008] A method for obtaining a character psychological portrait based on questionnaire evaluation, comprising:

[0009] Collect multiple dimensions of data from the psychological questionnaire assessment process;

[0010] Based on the multiple dimensional data of the evaluation process, deep learning is performed and collaborative clustering is performed to obtain the psychological portrait of the evaluator.

[0011] By adopting the above technical solution, on the basis of ordinary questionnaire evaluation, this application collects multiple dimensional data of the psychological questionnaire evaluation process, and conducts deep learning based on the multiple dimensional data of the evaluation process, and collaborative clustering to obtain the psychological portrait of the evaluator. Then, compared with simply conducting a questionnaire evaluation, the final psychological portrait of the person obtained is more accurate. At the same time, the above technical solution of this application has a wide range of social benefits and can fill the gap in the field of psychological evaluation that only focuses on results but lacks process data.

[0012] Preferably, the multiple dimensional data include one-dimensional data: number of option changes, duration of question dwelling, number of question replays; two-dimensional data: screen touch point coordinates and three-dimensional spatial data: acceleration.

[0013] By adopting the above technical solutions, one-dimensional, two-dimensional, and three-dimensional data are collected for comprehensive deep learning modeling, and ultimately a more accurate psychological portrait of the evaluated person is obtained.

[0014] Preferably, the multiple dimensional data are acquired in a silent manner, specifically including:

[0015] Read the number of questions, and set an entry timer array, an exit timer array, and an entry count counter array for each question;

[0016] Display the content of the question stem;

[0017] Read the number of options corresponding to the question, and set an option counter array and an option change times array for each option;

[0018] Display the option content and determine whether it is the last option of the current question;

[0019] If yes, then start the daemon process to record silently and collect multiple dimensional data of the user's psychological questionnaire evaluation process; otherwise, continue to display the option content until the options of the current question are displayed, then start the daemon process to record silently and collect multiple dimensional data of the user's psychological questionnaire evaluation process;

[0020] Determine whether the current question is the last question;

[0021] If so, after obtaining the user psychological questionnaire evaluation process data of the question, save the results of each counter and timer to obtain the multiple dimensional data; otherwise, go to the step of displaying the question content.

[0022] This application collects data in a silent manner. Compared with invasive data acquisition methods such as EEG, ECG, VR helmets, or passive (unnatural) data acquisition by the evaluator under camera surveillance, the collected data is more real and effective, which is conducive to ultimately obtaining a more accurate psychological portrait of the evaluated person.

[0023] Preferably, the questions are arranged with one question per page. Compared with the situation of multiple questions per page in the prior art (the estimated time is not accurate enough), the counter and timer of each question can count and time more accurately, collect more accurate data, and make the final psychological portrait of the person more accurate.

[0024] Preferably, before deep learning is performed, the method further includes performing data integration and normalization processing on the data; when processing the acceleration data, the following method steps are adopted:

[0025] Calculate the variance curve of the acceleration signal curve;

[0026] Performing data integration processing: segmenting the variance curve according to the period, and removing the interference band in the variance curve;

[0027] Normalization is performed: the acceleration signal value is converted from 0 to 1, and the mutation state is merged into the adjacent state.

[0028] By using the above technology for data processing, we can further obtain the answering posture - whether answering questions while walking, lying down, answering questions with shaking hands, etc., so as to better determine the answering mentality, and eliminate mutation data, so as to finally obtain a more accurate psychological portrait of the evaluated person.

[0029] Preferably, the collaborative clustering to obtain the psychological portrait of the evaluator includes:

[0030] The kmean algorithm was used to process the data of each evaluator;

[0031] Multiplying the processing result by a weight to amplify or reduce the processing result so that the result reflects the energy stage density of the collected data;

[0032] Find the minimum distance and the classification corresponding to the minimum distance;

[0033] Dynamically adjust the weight, and cyclically and repeatedly find the minimum distance and the classification corresponding to the minimum distance;

[0034] Obtaining the corresponding weight that makes the classification corresponding to the minimum distance more concentrated and the data of the evaluators outside the classification area the least as the final weight;

[0035] Based on the final weights, the final collaborative clustering results are obtained, and then the psychological portrait of the evaluator is obtained.

[0036] By adopting the above technical solution to obtain the psychological portrait of the evaluator,

[0037] The inventors found that when the existing algorithms are used for collaborative clustering, they usually use mean or variance to calculate data, but the mean will dilute the characteristics of the data in the high-energy stage; the variance will ignore vector characteristics such as the direction of motion. For example, in some robot algorithms or step recording algorithms, only variance data is used. However, the data of the motion sensor is positive or negative, and the data meaning will be lost after squaring. In other words, the variance method does not consider the motion characteristics of the data. Combining the above shortcomings of simply using mean and variance, this application adopts the kmean plus weighted scaling plus minimum distance algorithm, which has both the variance data of the motion sensor and the mean data. It can further analyze the posture of answering questions, whether walking to answer questions, lying down to answer questions, shaking hands to answer questions, etc., so as to better determine the answering mentality. At the same time, the weighted scaling and minimum distance methods can make the classification more concentrated, which is conducive to obtaining a more accurate psychological portrait of the evaluation character in the end.

[0038] Preferably, the multiple dimensional data correspond to multiple state indicators, and each state indicator corresponds to multiple feature vectors; the method also includes: determining whether the collected data matches the feature vector corresponding to a certain state indicator. If neither matches, the collected data is determined to be abnormal data, and the data is eliminated during deep learning modeling.

[0039] By adopting the above technical solution, abnormal data can be eliminated in advance, thereby improving the accuracy and efficiency of deep learning modeling.

[0040] In the second aspect, the present application provides a system for obtaining a character psychological portrait based on questionnaire evaluation, which adopts the following technical solution: A system for obtaining a character psychological portrait based on questionnaire evaluation, comprising:

[0041] Data collection module, used to collect multiple dimensions of data in the psychological questionnaire evaluation process;

[0042] The psychological portrait acquisition module is used to perform deep learning based on the multiple dimensional data and to obtain the psychological portrait of the evaluator through collaborative clustering.

[0043] In a third aspect, the present application provides an electronic device that adopts the following technical solution: an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that can be loaded and executed by the processor to implement any of the aforementioned methods.

[0044] In a fourth aspect, a computer-readable storage medium provided in the present application adopts the following technical solution: a computer-readable storage medium storing a computer program that can be loaded and executed by a processor to implement any of the aforementioned methods.

[0045] In summary, this application has the following beneficial technical effects:

[0046] By adopting the above technical solution, on the basis of ordinary questionnaire evaluation, this application collects multiple dimensional data of the psychological questionnaire evaluation process, and conducts deep learning based on the multiple dimensional data of the evaluation process, and collaborative clustering to obtain the psychological portrait of the evaluator. Then, compared with simply conducting a questionnaire evaluation, the final psychological portrait of the person obtained is more accurate. At the same time, the above technical solution of this application has a wide range of social benefits and can fill the gap in the field of psychological evaluation that only focuses on results but lacks process data. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a method flow chart of an embodiment of the present application.

[0048] Figure 2 This is a flow chart of a method for acquiring data in a silent manner in an embodiment of the present application.

[0049] Figure 3 It is a flow chart of a method for collaborative clustering to obtain a psychological portrait of an evaluator in an embodiment of the present application. DETAILED DESCRIPTION

[0050] The following is combined with Figure 1-Figure 3 This application is described in further detail.

[0051] The present application embodiment discloses a method for obtaining a character psychological portrait based on a questionnaire assessment. Figure 1 , a method for obtaining a character psychological portrait based on questionnaire evaluation, comprising:

[0052] S1, collects data from multiple dimensions of the psychological questionnaire assessment process;

[0053] S2, based on the multiple dimensional data of the evaluation process, deep learning is performed and collaborative clustering is performed to obtain the psychological portrait of the evaluator.

[0054] The portrait described above is to connect the user's psychology with the attributes and behaviors related to the questionnaire answers through data conversion, so as to determine the user's category.

[0055] Optionally, in step S1, the multiple dimensional data include one-dimensional data: number of option changes, duration of question dwell, number of question replays, two-dimensional data: screen touch point coordinates, and three-dimensional spatial data: acceleration.

[0056] The screen touch point coordinates are the two-dimensional coordinate values ​​of the screen touched by the finger during each question filling process. When the psychological assessment user submits the answer, the finger touches different screen coordinates. These screen coordinates reflect the user's answer track. The answer tracks of different users can be clustered to find similar individuals and discover their common problems. Similarly, the same psychological analysis is performed on individuals with the same track or the same clustering portrait, which not only saves manpower and material resources for intervention in the later stage of the assessment, but also greatly improves the efficiency of psychological services.

[0057] The three-dimensional data acceleration can be obtained through the acceleration sensor provided by the device.

[0058] In this embodiment, the multiple dimensional data are obtained in a silent manner, specifically including (such as Figure 2 shown):

[0059] Read the number of questions, and set an entry timer array, an exit timer array, and an entry count counter array for each question;

[0060] Display the content of the question stem;

[0061] Read the number of options corresponding to the question, and set an option counter array and an option change times array for each option;

[0062] Display the option content and determine whether it is the last option of the current question;

[0063] If yes, start the daemon process to record silently and collect multiple dimensional data of the user's psychological questionnaire evaluation process; otherwise, continue to display the option content until the options of the current question are displayed, start the daemon process to record silently and collect multiple dimensional data of the user's psychological questionnaire evaluation process (mainly collect two-dimensional and three-dimensional data);

[0064] Determine whether the current question is the last question;

[0065] If so, after obtaining the user psychological questionnaire evaluation process data of the question, save the results of each counter and timer (obtain one-dimensional data), and then obtain the multiple dimensional data; otherwise, go to the step of displaying the question content.

[0066] Optionally, during specific implementation, you can first determine the type of question, distinguish between single-choice, multiple-choice, voting and other types, which will affect the calculation of the number of times the options are changed.

[0067] In other embodiments, the multiple dimensional data may also be acquired by invasive data acquisition methods such as EEG, ECG, VR helmets, etc., or by the evaluator in a passive manner (unnatural manner) under camera surveillance.

[0068] In specific implementation, the duration of one entry to answer a question can be obtained by subtracting the entry timer and exit timer values ​​of each question; and combined with the entry count counter array, the total duration of multiple entries to answer the question can be calculated.

[0069] When counting the number of option changes, each of the option counter arrays and option change number arrays can be used. When a certain option is changed, the corresponding option change number array is incremented by one; then the total number of question changes is also incremented by one.

[0070] Optionally, after step S1 collects data, a series of preprocessing can be performed first, such as deleting invalid data, filling missing fields, cleaning up duplicate data when processing raw data, and then selecting data for the screened attributes, establishing a data set, and finally converting different types of data into data that can be processed uniformly, such as converting male gender into data 1 and female gender into data 0; recording "yes" and "no" text answers as 1 and 0, etc.

[0071] In order to pre-process the collected data in different dimensions in a unified manner, the collected data can be formed into different databases. For example:

[0072] Analyze and extract data from the psychological assessment process, then clean the data, sort out and integrate data such as option changes and answering time, and form an answer database (i.e., a one-dimensional data database) through feature engineering methods;

[0073] Perform RPX unification on the touch point coordinates in the process data, map the relative positions, and then perform statistical summary analysis to form a touch point coordinate database (i.e., a database of two-dimensional data) for each questionnaire process data;

[0074] The acceleration sensor data is extracted from the evaluation process, extreme values ​​and other data are removed, the variance and mean are calculated, and the action database (i.e., the three-dimensional data database) in the questionnaire process data is formed.

[0075] In this embodiment, the questions are set to one question per page. In other embodiments, multiple questions per page may also be set.

[0076] Optionally, in step S2, before deep learning is performed, the data may be integrated and normalized; when the acceleration data is processed, the following method steps may be used:

[0077] Calculate the variance curve of the acceleration signal curve;

[0078] Performing data integration processing: segmenting the variance curve according to the period, and removing the interference band in the variance curve;

[0079] Normalization is performed: the acceleration signal value is converted from 0 to 1, and the mutation state is merged into the adjacent state.

[0080] In this embodiment, in step S2, the collaborative clustering to obtain the psychological portrait of the evaluator includes (e.g. Figure 3 shown):

[0081] S21, the kmean algorithm was used to process the data of each evaluator;

[0082] S22, multiplying the processing result by a weight to amplify or reduce the processing result so that the result reflects the energy stage density of the collected data;

[0083] S23, finding the minimum distance and the category corresponding to the minimum distance;

[0084] S24, dynamically adjusting the weight, and cyclically and repeatedly finding the minimum distance and the classification corresponding to the minimum distance;

[0085] S25, obtaining the corresponding weight that makes the classification corresponding to the minimum distance more concentrated and the data of the evaluators outside the classification area (the data outside the clustering result is the abnormal data) the least as the final weight;

[0086] S26, based on the final weight, obtaining the final collaborative clustering result, and then obtaining the psychological portrait of the evaluator.

[0087] The dynamic adjustment of the weight may adjust the weight value from small to large; the preferred range may be 0.6-1.5.

[0088] In other embodiments, the mean or variance may also be used to calculate the final co-clustering result, thereby obtaining the psychological portrait of the evaluator.

[0089] Optionally, the multiple dimensional data correspond to multiple state indicators, and each state indicator corresponds to multiple feature vectors; the deep learning based on the multiple dimensional data of the evaluation process described in step S2 also includes: the multiple dimensional data correspond to multiple state indicators, and each state indicator corresponds to multiple feature vectors; the method also includes: determining whether the collected data matches the feature vector corresponding to a certain state indicator, and if neither matches, determining that the collected data is abnormal data, and excluding the data during deep learning modeling.

[0090] Specifically, if a set of feature vectors is matched, the data corresponds to the corresponding category of the relevant feature vectors. If the data does not match any of the current categories, it is classified as abnormal data and can be left for manual screening, and the manual annotation results are used as sensitive data for future deep learning modeling. For example, for the answering time indicator, according to the answering time proportion, the extremely short answering proportion, the distracted state proportion, etc., after data clustering and combining the feature vectors of different user tests, one-dimensional time data can derive multiple sets of feature vectors. When the newly collected time data matches the feature vector, four types of answering time feature vectors belonging to normal, over-concern, under-concern, and abnormal can be obtained. For the option change indicator, after clustering and matching the feature vectors, the one-dimensional time feature vector and the two-dimensional contact coordinate feature vector are combined to obtain four types of option change types: early question review change, mid-term deduction change, late wrong selection change, and abnormal. Similarly, for the three-dimensional spatiotemporal data of the acceleration sensor, after clustering and matching the feature vectors, four feature vectors and corresponding indicators of normal, anxious, distracted, and abnormal can be obtained.

[0091] The embodiment of the present application also discloses a system for obtaining a character psychological portrait based on a questionnaire assessment. A system for obtaining a character psychological portrait based on a questionnaire assessment includes:

[0092] Data collection module, used to collect multiple dimensions of data in the psychological questionnaire evaluation process;

[0093] The psychological portrait acquisition module is used to perform deep learning based on the multiple dimensional data and to obtain the psychological portrait of the evaluator through collaborative clustering.

[0094] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0095] The embodiment of the present application also discloses an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that can be loaded and executed by the processor to implement any of the above methods.

[0096] The electronic device may be a mobile phone, a tablet or other electronic device, and the electronic device includes but is not limited to a processor and a memory. For example, the electronic device may also include input and output devices, a network access device and a bus.

[0097] The processor in the present application may include one or more processing cores. The processor calls the data stored in the memory by running or executing the instructions, programs, code sets or instruction sets stored in the memory, and executes the various functions and processing data of the present application. The processor can be at least one of an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a central processing unit (CPU), a controller, a microcontroller and a microprocessor. It is understandable that for different devices, the electronic device used to realize the above-mentioned processor function can also be other, and the embodiments of the present application are not specifically limited.

[0098] Among them, the memory can be an internal storage unit of the electronic device, such as a hard disk or memory of the electronic device, or it can be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (SMC), a secure digital card (SD) or a flash memory card (FC) equipped on the electronic device. Moreover, the memory can also be a combination of an internal storage unit and an external storage device of the electronic device. The memory is used to store computer programs and other programs and data required by the electronic device. The memory can also be used to temporarily store data that has been output or is to be output, and this application does not impose any restrictions on this.

[0099] The embodiment of the present application also discloses a computer-readable storage medium. A computer-readable storage medium stores a computer program that can be loaded and executed by a processor to implement any of the above methods.

[0100] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0101] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, all equivalent changes made according to the methods and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for obtaining a character psychological portrait based on questionnaire evaluation, characterized in that: include: Collect multiple dimensional data of the psychological questionnaire evaluation process; the multiple dimensional data include one-dimensional data: number of option changes, duration of question stay, number of question review; Two-dimensional data: screen touch point coordinates and three-dimensional spatial data: acceleration; Based on the multiple dimensional data of the evaluation process, deep learning is performed and collaborative clustering is performed to obtain the psychological portrait of the evaluator; The multiple dimensional data are acquired in a silent manner, specifically including: Read the number of questions, and set an entry timer array, an exit timer array, and an entry times counter array for each question; the questions are set to one question per page; Display the content of the question stem; Read the number of options corresponding to the question, and set an option counter array and an option change times array for each option; Display the option content and determine whether it is the last option of the current question; If yes, then start the daemon process to record silently and collect multiple dimensional data of the user's psychological questionnaire evaluation process; otherwise, continue to display the option content until the options of the current question are displayed, then start the daemon process to record silently and collect multiple dimensional data of the user's psychological questionnaire evaluation process; Determine whether the current question is the last question; If yes, after obtaining the user psychological questionnaire evaluation process data of the question, save the results of each counter and timer, and then obtain the multiple dimensional data; otherwise, go to the step of displaying the question content; The collaborative clustering to obtain the psychological portrait of the evaluator includes: The kmean algorithm was used to process the data of each evaluator; Multiplying the processing result by a weight to amplify or reduce the processing result so that the result reflects the energy stage density of the collected data; Find the minimum distance and the classification corresponding to the minimum distance; Dynamically adjust the weight, and cyclically and repeatedly find the minimum distance and the classification corresponding to the minimum distance; Obtaining the corresponding weight that makes the classification corresponding to the minimum distance more concentrated and the data of the evaluators outside the classification area the least as the final weight; Based on the final weights, the final collaborative clustering results are obtained, and then the psychological portrait of the evaluator is obtained.

2. The method for obtaining a character psychological portrait based on questionnaire evaluation according to claim 1, characterized in that: Before deep learning is performed, data integration and normalization are also performed on the data; when processing the acceleration data, the following method steps are adopted: Calculate the variance curve of the acceleration signal curve; Performing data integration processing: segmenting the variance curve according to the period, and removing the interference band in the variance curve; Normalization is performed: the acceleration signal value is converted from 0 to 1, and the mutation state is merged into the adjacent state.

3. The method for obtaining a character psychological portrait based on questionnaire evaluation according to claim 1, characterized in that: The multiple dimensional data correspond to multiple state indicators, and each state indicator corresponds to multiple feature vectors; the method also includes: determining whether the collected data matches the feature vector corresponding to a certain state indicator. If neither matches, the collected data is determined to be abnormal data, and the data is eliminated during deep learning modeling.

4. A character psychological portrait acquisition system based on questionnaire evaluation, characterized in that: include: The data collection module is used to collect multiple dimensional data of the psychological questionnaire evaluation process; the multiple dimensional data include one-dimensional data: the number of option changes, the length of time the question stays, and the number of times the question is reviewed; Two-dimensional data: screen touch point coordinates and three-dimensional spatial data: acceleration; A psychological portrait acquisition module, used for performing deep learning based on the multiple dimensional data and collaborative clustering to obtain a psychological portrait of the evaluator; Wherein, the data acquisition module comprises: The first setting submodule is used to read the number of questions, and set an entry timer array, an exit timer array and an entry times counter array for each question; the questions are set to one question per page; Display submodule, used to display the content of the question; The second setting submodule is used to read the number of options corresponding to the question, and set an option counter array and an option change times array for each option; The display and judgment submodule is used to display the option content and judge whether it is the last option of the current question; If yes, execute the data collection submodule to start the daemon process to record silently and collect multiple dimensions of data during the user psychological questionnaire evaluation process; If not, the display and recording submodule is executed to continue to display the option content until the options of the current question are displayed, and then the daemon process is started to record silently to collect multiple dimensional data of the user psychological questionnaire evaluation process; The judgment submodule is used to judge whether the current question is the last question; If yes, execute the data saving submodule to obtain the user psychological questionnaire evaluation process data of the question, save the results of each counter and timer, and then obtain the multiple dimensional data; otherwise, go to the step of displaying the question content; The psychological portrait acquisition module includes: The data processing submodule is used to process the data of each evaluator using the kmean algorithm; A weight setting submodule, used to multiply the processing result by a weight, and to amplify or reduce the processing result so that the result reflects the energy stage density of the collected data; A minimum distance and classification acquisition submodule is used to obtain the minimum distance and the classification corresponding to the minimum distance; A weight adjustment submodule, used to dynamically adjust the weight, and cyclically and repeatedly find the minimum distance and the classification corresponding to the minimum distance; A weight acquisition submodule is used to obtain the corresponding weight that makes the classification corresponding to the minimum distance more concentrated and the data of the evaluators outside the classification area the least as the final weight; The psychological portrait acquisition submodule is used to obtain the final collaborative clustering result based on the final weight, and then obtain the psychological portrait of the evaluator.

5. An electronic device, characterized in that: The invention comprises a memory and a processor, wherein the memory stores a computer program which can be loaded and executed by the processor to implement any one of the methods in claims 1 to 3.

6. A computer-readable storage medium, characterized in that: A computer program capable of being loaded and executed by a processor to implement any one of the methods of claims 1 to 3 is stored.

Images