A psychological state assessment method and device based on an artificial intelligence algorithm model

Abstract

The application discloses a psychological state evaluation method and device based on an artificial intelligence algorithm model, and belongs to the technical field of psychological state evaluation, and comprises the following steps: a tester inputs text information into an emotional feature recognition algorithm model; the emotional feature recognition algorithm model analyzes the text information of the tester, and extracts psychological state related features; the obtained psychological state related features are matched with a psychological state feature library to obtain a matching result; according to the matching result, a personalized evaluation question group is generated; and after a user completes the personalized evaluation question group, a psychological state evaluation result is obtained. Through human-computer conversation, the emotional and psychological reactions of the tested person in natural conversation are captured in real time, and the inaccuracy caused by traditional scales is avoided. The natural conversation mode can more truly reflect the inner state of the tested person. The AI algorithm can perform real-time analysis through conversation, dynamically adjust the evaluation content according to the reactions of the tested person, reduce the emotional influence, and make the evaluation result more comprehensive and objective.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of psychological state assessment, and in particular to a psychological state assessment method and device based on an artificial intelligence algorithm model. BACKGROUND

[0002] Psychological state scales are an important tool in the field of mental health. Commonly used psychological scales include the State Self-Assessment Scale (SCL-90), the Beck Depression Inventory (BDI), the Anxiety Self-Assessment Scale (SAS), and the Health Lifestyle Scale, etc. Psychological state assessment scales are mostly based on large sample psychological data, and have undergone a standardized development process, which can provide high-comparability assessment results for different populations.

[0003] However, the scales widely used for psychological state assessment currently have some problems in practice, which can affect the accuracy and reliability of the assessment, such as: 1) test fatigue: long-term testing leads to fatigue of the testers, affecting the quality of their answers to the questions; 2) social desirability bias: the subjects will answer according to the answers they think the society or the assessor expects, rather than honestly reflecting their actual feelings or state; 3) self-image maintenance; 4) emotional fluctuation influence: the subjects' immediate emotions when filling out the scale can also affect their answers; 5) avoidance of negative experiences; 6) deliberately exaggerating or minimizing the state; etc.

[0004] In order to reduce the influence of the above problems, the present application uses artificial intelligence technology to optimize the assessment of psychological state, providing more personalized and accurate assessment, while introducing a tool to monitor the dynamic changes in psychological state.

[0005] In current psychological state assessment, paper-based scales or paper-based scales that have been digitized (such as electronic forms or simple electronic forms) are still the mainstream assessment tools;

[0006] 1) Lack of individualized and dynamic assessment capabilities

[0007] Paper or simple electronic scales are usually designed as standardized questionnaires, which are suitable for the general public but lack individualized design; they cannot track the changes in individual psychological state in real time and cannot provide long-term, continuous assessment data. Psychological state has the characteristics of fluctuation and dynamic change, and the results of a single assessment are often insufficient to reflect the true psychological state of the subjects.

[0008] 2) Lack of intelligent and data integration capabilities

[0009] Paper and simple electronic forms rely on traditional manual scoring and analysis, making it difficult to use modern technologies such as artificial intelligence and machine learning for intelligent data analysis and pattern recognition. This makes it difficult to handle large-scale and diverse psychological data. It is unable to integrate other types of data (such as psychological data, behavioral data, etc.) for comprehensive analysis. This single data source limits the comprehensiveness and accuracy of psychological assessment.

[0010] 3) Subjectivity problem

[0011] Paper and simple electronic forms rely on self-reporting by the testee, which is influenced by emotions, motivation and cognitive ability. The testee's answers may not be accurate or true, leading to biased assessment results. This problem is particularly evident in static assessment tools that lack intelligent feedback and supervision. Testees may give different answers to the same question at different times and in different situations. Traditional paper or simple electronic forms cannot identify these inconsistencies and cannot provide real-time intervention or correction.

[0012] 4) Lack of immediacy and interactivity in assessment

[0013] Paper and simple electronic forms usually cannot provide immediate feedback or results, and the testee needs to wait for the assessment to be completed before obtaining the analysis results. For mental health problems that require immediate intervention or continuous monitoring, this lag may delay timely treatment of the problem.

[0014] These scales lack interaction with the testee, and cannot dynamically adjust the difficulty or number of questions based on the testee's answers, nor can they provide personalized assessment experiences based on individual differences. This limits the application of the scale in complex psychological assessment. SUMMARY

[0015] The purpose of the present application is to provide a psychological state assessment method and device based on artificial intelligence algorithm model to overcome the shortcomings of the prior art.

[0016] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0017] The present application discloses a psychological state assessment method based on artificial intelligence algorithm model, comprising the following steps:

[0018] S1, the testee inputs the text information into the emotion feature recognition algorithm model;

[0019] S2, the emotion feature recognition algorithm model analyzes the testee's text information and extracts psychological state related features;

[0020] S3, match the psychological state related features obtained in step S2 with the psychological state feature library to obtain a matching result; according to the matching result, generate a personalized assessment question group;

[0021] S4, after the user completes the personalized evaluation question group of step S3, the psychological state evaluation result is obtained.

[0022] Preferably, in step S1, the tester inputs the content that makes him feel disturbed or uncomfortable as text information through text or voice.

[0023] Preferably, step S2 specifically comprises the following sub-steps:

[0024] S21, the text information is standardized to obtain preprocessed text; the preprocessed text is input into a GPT model to generate a high-dimensional semantic embedding vector; the semantic information and context dependency in the preprocessed text are captured through the high-dimensional semantic embedding vector, and the emotion-related semantic features are retained;

[0025] S22, in combination with the emotion-related semantic features and emotional vocabulary obtained in step S21, the closely related words, phrases and context relationships in the preprocessed text are identified, and the emotion-related features are extracted;

[0026] S23, the emotion-related features obtained in step S22 and the high-dimensional semantic embedding vector are input into a multi-level emotion classifier to extract psychological state-related features; the multi-level emotion classifier is used to classify the emotion in the text.

[0027] Preferably, in step S21, the standardization processing includes removing irrelevant characters and stop words.

[0028] Preferably, in step S23, the multi-level emotion classifier adopts a deep learning model, including a classifier of Transformer architecture.

[0029] Preferably, in step S3, the psychological state feature library is established in the following manner:

[0030] S31, based on the psychological state self-evaluation scale, according to the several items in each dimension of the psychological state self-evaluation scale that describe specific psychological states, the items are mapped into feature vectors of emotions and psychological states, and specific state descriptions under each dimension are formed;

[0031] S32, mapping the specific state description under each dimension to the corresponding emotion category.

[0032] Preferably, in step S4, according to the personalized evaluation question group completed by the user, the evaluation result is output in combination with the evaluation method of the psychological state self-evaluation scale.

[0033] The application also discloses a psychological state evaluation device based on an artificial intelligence algorithm model, comprising a memory and one or more processors, the memory stores executable code, and the one or more processors execute the executable code to implement the psychological state evaluation method based on the artificial intelligence algorithm model.

[0034] The application also discloses a computer-readable storage medium, which stores a program, and the program is executed by a processor to implement the psychological state evaluation method based on the artificial intelligence algorithm model.

[0035] The application has the following advantages:

[0036] 1) Reducing subjectivity and reporting bias: By real-time capturing the emotions and psychological reactions of the subjects in natural conversation through human-computer dialogue, the inaccuracy caused by self-reporting bias, social desirability effect and other problems in traditional scales is avoided. This natural conversation method can more truly reflect the inner state of the subjects. The AI algorithm can analyze the conversation in real time, dynamically adjust the evaluation content according to the reactions of the subjects, reduce the influence of emotions at a single time point, and make the evaluation results more comprehensive and objective.

[0037] 2) Improving evaluation efficiency: The system can automatically analyze the psychological state in the conversation, saving the steps of answering questions one by one, manual filling and evaluation required by traditional scales. The AI model can predict the state in a short time and quickly match suitable psychological evaluation questions, thereby improving the speed and efficiency of evaluation. After the AI detects potential problems, it can quickly guide the subjects to conduct more detailed psychological tests or recommend the next step of intervention measures, reduce manual operation and improve the efficiency of clinical practice.

[0038] 3) Personalization and dynamic evaluation: Based on the preliminary state analysis of AI, the system can dynamically adjust the types and sequences of questions according to the specific performance of individuals. Compared with the standardized design of traditional scales, the application can provide personalized evaluation paths for each subject and capture more targeted psychological problems.

[0039] The features and advantages of the application will be described in detail in conjunction with the embodiments and the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 is a flowchart of the psychological state evaluation method based on the artificial intelligence algorithm model of the application;

[0041] Figure 2 is a schematic diagram of efficiency evaluation of the application and conventional tests;

[0042] Figure 3 is a schematic diagram of accuracy evaluation of the application and traditional state scales;

[0043] Figure 4 is a structural schematic diagram of a psychological state evaluation device based on an artificial intelligence algorithm model DETAILED DESCRIPTION

[0044] To make the objectives, technical solutions, and advantages of the present application clearer and more comprehensible, the present application is further described in detail below with the aid of drawings and examples. However, it should be understood that the specific examples described herein are merely used to explain the present application and are not used to limit the scope of the present application. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily confusing the concept of the present application.

[0045] Referring to Figure 1 The psychological state evaluation method based on the artificial intelligence algorithm model is specifically operated as follows:

[0046] At the beginning of the test, the user can input the content that recently makes him or her feel troubled or uncomfortable through text or voice as the start of the conversation. The built-in emotion feature recognition algorithm of the system will automatically analyze the user input and extract features related to the psychological state. Subsequently, these features will be matched with the psychological state feature library, and the system will generate personalized evaluation questions for the user according to the matching results. After the user completes these personalized test questions in the conversation box, the system generates a psychological state evaluation result in real time, realizing accurate evaluation of the psychological state.

[0047] Regarding the construction process of the emotion feature recognition algorithm model:

[0048] The emotion feature recognition algorithm model combines the GPT4.0 large language model and the sentiment classifier, aiming to accurately and efficiently recognize the emotional features in the input text. The specific steps are as follows:

[0049] (1) Text preprocessing and embedding generation: The input text of the tester is standardized, including removing irrelevant characters and stop words. The processed text is input into the GPT4.0 model to generate high-dimensional semantic embedding vectors. These embedding vectors can capture deep semantic information and context dependencies in the text, thereby preserving emotion-related language features.

[0050] (2) Emotion feature extraction: Based on the generated text embedding, further extract emotion-related features. Combine semantic features and emotion vocabulary to identify words, phrases, and their context relationships closely related to emotions in the text.

[0051] (3) Multi-level sentiment classifier: The extracted sentiment features and language embedding vectors are input into a multi-level sentiment classifier. This classifier uses a deep learning model (Transformer architecture classifier) that, after training, can classify emotions in text. This classifier not only distinguishes basic sentiment categories in text (such as positive, negative, and neutral), but also identifies complex emotions (such as the coexistence of anger and disappointment) through a multi-label learning mechanism.

[0052] (4) Model optimization and fine-tuning: During the construction process, the sentiment classifier is trained on a large number of labeled data sets (such as the public data set GoEmotions emotion data set and the self-built psychological feature data set).

[0053] Regarding the establishment of the psychological state feature library:

[0054] The Psychological State Self-Rating Scale (SCL-90) is a widely used tool for assessing individual psychological states, covering 9 main dimensions. Based on this scale, a psychological state feature library is established, with the following specific process:

[0055] (1) State dimension definition: Each dimension contains several items describing specific psychological states. These items are mapped to emotion and psychological state feature vectors to form specific state descriptions under each dimension. (2) Feature labeling and classification: The psychological states of the 9 dimensions are managed with corresponding emotion features. The state of each dimension is mapped to the corresponding emotion category.

[0056] Regarding the automatic generation process of the individualized assessment question set:

[0057] After the testee completes the information input, the emotion feature recognition algorithm model extracts the emotion features. The extracted features are matched with the psychological state feature library to obtain the psychological state features that appear in the testee's communication process (such as the model locating to the interpersonal sensitivity and depression features), and the system will extract the questions related to the interpersonal sensitivity and depression feature dimensions in the psychological state self-rating scale (SCL-90) and give them to the testee for evaluation. The evaluation results are output according to the evaluation method of the psychological state self-rating scale (SCL90).

[0058] The psychological state evaluation method based on artificial intelligence algorithm model is based on artificial intelligence algorithm model, which captures the psychological state reflected in the conversation of the testee through human-computer conversation, realizes the pre-judgment of the psychological state problem of the testee, and then matches the psychological evaluation test questions based on the state of algorithm analysis to confirm the psychological state of the testee.

[0059] The implementation of the present application can solve the problems of current paper quality table or simple electronic scale 1) lack of individualization and dynamic evaluation capability: the algorithm model automatically matches the psychological test questions related to the testee by capturing the key psychological state information in the testee's text or language communication; and the system generates dynamic tracking data for the test results of the testee, realizing long-term dynamic state evaluation and management of the testee. 2) Lack of intelligent and data integration capability: the present application based on artificial intelligence algorithm model can realize real-time calculation and analysis of evaluation content, and has an external interface, which can be connected and integrated with brain electrical equipment data, realizing comprehensive evaluation of psychological state. 3) Subjectivity problem: the algorithm model of the present application can mine the psychological state of the testee through text or voice communication with the testee, solving the problem that the result accuracy of the conventional paper evaluation scale is affected by the subjectivity of the testee. 4) Lack of immediacy and interactivity of evaluation: the evaluation system of the present application can be integrated in WeChat or used in the form of a separate app, and the testee can use it at any time. At the same time, the system can automatically reply in real time based on the input content of the testee, meeting the immediacy and interactivity requirements of the testee for psychological state evaluation.

[0060] Referring to Figure 2 The test efficiency evaluation shows that, taking SCL90 as an example, the average time for completing the test in the system is 7.8 minutes, which is about 1 / 3 of the time for conventional test.

[0061] Referring to Figure 3 , the system test accuracy evaluation shows that, compared with the traditional state scale screening, the overall accuracy of the screening results of the present application is improved by 20%.

[0062] The embodiment of the present application can be applied to any device with data processing capability, such as a computer or the like. The device embodiment can be realized by software, or by hardware or a combination of software and hardware. Taking software realization as an example, as a logical device, it is formed by reading the corresponding computer program instructions in the non-volatile memory into the memory and running through the processor of the device with data processing capability. From the hardware level, as shown in Figure 4 , it is a hardware structure diagram of the device with data processing capability where the present application is located. In addition to Figure 4In addition to the processor, the memory, the network interface, and the nonvolatile memory, any data processing capable device in which the apparatus in the embodiments is located can also include other hardware according to the actual functions of the data processing capable device, and details are not described herein. The implementation process of the functions and roles of each unit in the apparatus is specifically described in the implementation process of the corresponding steps in the above method, and details are not described herein.

[0063] For the apparatus embodiments, since they basically correspond to the method embodiments, the relevant parts are described in the part of the method embodiments. The apparatus embodiments described above are only illustrative, and the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, that is, they can be located in one place or distributed on multiple network units. Some or all of the modules can be selected to achieve the purpose of the present application according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0064] The embodiments of the present application also provide a computer readable storage medium, which stores a program, and the program is executed by a processor to realize the psychological state evaluation apparatus based on the artificial intelligence algorithm model in the above embodiments.

[0065] The computer readable storage medium can be an internal storage unit of any data processing capable device, such as a hard disk or a memory. The computer readable storage medium can also be an external storage device of any data processing capable device, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. Further, the computer readable storage medium can include both the internal storage unit and the external storage device of any data processing capable device. The computer readable storage medium is used to store the computer program and other programs and data required by the data processing capable device, and can also be used to temporarily store data that has been output or will be output.

[0066] The above only describes the preferred embodiments of the present application and does not limit the present application. Any modification, equivalent replacement or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A mental state evaluation method based on an artificial intelligence algorithm model, characterized in that, The method comprises the following steps: S1, the tester inputs the text information into the emotional feature recognition algorithm model; S2, the emotional feature recognition algorithm model analyzes the text information of the tester, and extracts the psychological state related features; S3, match the psychological state related features obtained in step S2 with the psychological state feature library to obtain a matching result; generate a personalized assessment question group according to the matching result; S4, after the user completes the personalized assessment question group in step S3, the psychological state assessment result is obtained; In step S1, the tester inputs the content that makes him feel troubled or uncomfortable as text information through text or voice input; Step S2 specifically comprises the following substeps: S21, standardize the text information to obtain preprocessed text; input the preprocessed text into the GPT model to generate high-dimensional semantic embedding vectors; capture semantic information and context dependency in the preprocessed text through high-dimensional semantic embedding vectors, and retain emotion-related semantic features; S22, combine the emotion-related semantic features obtained in step S21 and the emotional vocabulary to identify the words, phrases and context relationships closely related to emotions in the preprocessed text, and extract emotion-related features; S23, input the emotion-related features obtained in step S22 and the high-dimensional semantic embedding vectors into the multi-level emotion classifier to extract the psychological state related features; the multi-level emotion classifier is used to classify the emotions in the text; The establishment method of the psychological state feature library in step S3 is as follows: S31, based on the psychological state self-assessment scale, according to the several items in the psychological state self-assessment scale that describe specific psychological states, map them into emotional and psychological state feature vectors to form specific state descriptions under each dimension; S32, map the specific state descriptions under each dimension to the corresponding emotional categories; The multi-level emotion classifier: a classifier using the deep learning model Transformer architecture, which can classify emotions in text after training; this classifier can not only distinguish the basic emotional categories in the text, but also identify complex emotions through a multi-label learning mechanism. 2.The psychological state evaluation method based on the artificial intelligence algorithm model according to claim 1, wherein In step S21, the standardization process includes removing irrelevant characters and stop words. 3.The psychological state evaluation method based on the artificial intelligence algorithm model of claim 1, wherein, In step S4, according to the personalized assessment question group completed by the user, the evaluation result is output in combination with the evaluation method of the psychological state self-assessment scale.

4. A mental state evaluation device based on an artificial intelligence algorithm model, characterized by: The memory and one or more processors, the memory has stored executable code, the one or more processors execute the executable code to implement the psychological state assessment method based on the artificial intelligence algorithm model in any one of claims 1-3.

5. A computer-readable storage medium, characterized in that: A program is stored thereon, which is executed by a processor to implement the psychological state assessment method based on the artificial intelligence algorithm model in any one of claims 1-3.

Images