Teenager mental health management system based on knowledge base and AI

Abstract

The invention discloses a teenager mental health management system based on a knowledge base and AI, and relates to the field of artificial intelligence and psychology. Data including input, daily interaction, psychological evaluation and other sources of teenager users are collected, the collected data are analyzed through a large model, text input of the teenagers is analyzed through a natural language processing technology, emotion and emotional states of the input users are recognized, and the psychological health requirements of the teenagers are obtained. Defining at least one task corresponding to the psychological health demand according to the identification result of the psychological health demand, setting the priority of the task, dynamically allocating the task to a corresponding service provider by using an AI technology, and adjusting the execution sequence and content of the task according to the real-time condition, tasks are distributed to corresponding execution units through a task arrangement engine, whole-process monitoring is carried out, user feedback in the task execution process is obtained, the user feedback is analyzed in real time through a large model, and instant adjustment suggestions are provided.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence and psychology, and in particular to a youth mental health management system based on a knowledge base and AI. Background Art

[0002] In recent years, artificial intelligence technology has been widely used in the field of medical health, especially in mental health management. Through technologies such as big data analysis, machine learning, and deep learning, mental health problems can be more accurately assessed and intervened. However, the current intelligent system for adolescent mental health is not mature enough and has the following defects: 1) The existing system has limited capabilities in mental health data analysis and cannot provide in-depth mental state assessment; 2) The mental health data in the existing system is scattered on different platforms, making it difficult to manage and analyze uniformly; 3) The existing system lacks an effective service scheduling mechanism and cannot dynamically adjust the provision of mental health services according to individual needs. To this end, we propose a youth mental health management system based on knowledge base and AI. Summary of the invention

[0003] The main purpose of the present invention is to provide a youth mental health management system based on knowledge base and AI, which can effectively solve the problems in the background technology.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] The adolescent mental health management system based on knowledge base and AI, the implementation process of the system includes the following steps:

[0006] Collect data including input from adolescent users, daily interactions, psychological assessments, and other sources;

[0007] Use big models to analyze the collected data to obtain the mental health needs of adolescents.

[0008] By using natural language processing technology to analyze teenagers' text input, identify the user's emotions and emotional state, and adopt an emotion-aware optimization model, EPOM combines LSTM and Attention mechanisms, and makes full use of the sentiment analysis capabilities of large models, making it more accurate and efficient in identifying users' emotions and emotional states;

[0009] In the specific implementation process, the EPOM model first encodes the input text through the LSTM network to capture the contextual information and emotional changes in the text. Subsequently, the Attention mechanism performs weighted processing on the encoded information and dynamically focuses on the important parts of the text, thereby improving the accuracy of emotion recognition. Finally, through the sentiment analysis capabilities of the large model and the output results of the EPOM model, a comprehensive assessment of the user's emotions and emotional state is carried out.

[0010] This optimized algorithm model can not only significantly improve the accuracy of user emotion and emotional state recognition, but also speed up the recognition process, providing more accurate support for subsequent mental health management services.

[0011] Based on the results of the identification of mental health needs, define at least one task corresponding to the mental health needs and set the priority of the tasks;

[0012] Use AI technology to dynamically schedule tasks, assign tasks to corresponding service providers, and adjust the execution order and content of tasks according to real-time conditions:

[0013] The task scheduling engine allocates tasks to corresponding execution units and monitors the entire process.

[0014] Obtain user feedback during the task execution process, analyze user feedback in real time through a large model, and provide immediate adjustment suggestions.

[0015] The implementation process of the system also includes the following steps:

[0016] Construct a youth mental health data knowledge base and dynamically update the knowledge base, wherein the knowledge base at least includes classified data on youth mental health needs, standard processes and recommended data for all tasks, best practices and reference data for all task executions, execution progress and effect data for all tasks, and the latest research and practical experience data on youth mental health.

[0017] The knowledge base is dynamically updated using the following methods:

[0018] Method 1: System developers manually input the latest research results, standard processes, and reference data into the knowledge base to update the knowledge base content;

[0019] Method 2: Collect question and answer data between users and the system, and update the knowledge base in real time based on the obtained question and answer data.

[0020] The implementation process of the system also includes the following steps:

[0021] An upper limit value of the question-and-answer data of the user and the system is set. When the question-and-answer data accumulates to the upper limit value, the acquired question-and-answer data is learned through the big model to update the big model.

[0022] The specific process includes the following steps:

[0023] Collecting question and answer data, where the sources of the question and answer data include external input and system log records;

[0024] Data cleaning and labeling: When the amount of question and answer data reaches the upper limit, the collected question and answer data will be preprocessed;

[0025] Model fine-tuning, using the preprocessed question-answering data to fine-tune the large model.

[0026] The implementation process of question-answer data preprocessing includes the following steps:

[0027] The collected question and answer data is deduplicated through similarity algorithms;

[0028] Set keywords and use them to filter and clean the collected question and answer data;

[0029] Annotate the collected question-answering data.

[0030] The system includes a data acquisition module, a large model module, a knowledge base, a task scheduling engine module, and a user feedback acquisition module;

[0031] The data collection module is used to collect data including input from adolescent users, daily interactions, psychological assessments and other sources;

[0032] The large model module is used to identify the emotional and emotional state of the input user, obtain the mental health needs of the adolescent, define at least one corresponding task according to the mental health needs, set the priority of the task, dynamically schedule the task, and assign the task to the corresponding service provider:

[0033] The task scheduling engine module is used to assign tasks to corresponding execution units and monitor the entire process;

[0034] The user feedback acquisition module is used to obtain user feedback during the execution of the task, analyze the user feedback in real time through the big model, and provide instant adjustment suggestions.

[0035] The large model module includes a mental health needs analysis submodule, a task definition submodule, a task scheduling submodule, and a model update submodule.

[0036] The mental health needs analysis submodule is used to analyze the collected data, analyze the text input of the teenagers through natural language processing technology, identify the emotions and emotional states of the input users, and obtain the mental health needs of the teenagers;

[0037] The task definition submodule is used to define at least one task corresponding to the mental health need according to the identification result of the mental health need, and set the priority of the task;

[0038] The task scheduling submodule uses AI technology to dynamically schedule tasks, assign tasks to corresponding service providers, and adjust the execution order and content of tasks according to real-time conditions:

[0039] The model updating submodule is used to preprocess the collected question and answer data, learn from the preprocessed question and answer data, and fine-tune the large model according to the learning results.

[0040] The system also includes a knowledge base dynamic update module for updating the knowledge base content, and the knowledge base dynamic update module includes a data input submodule, a question and answer data collection submodule, and a question and answer data processing submodule;

[0041] The data input submodule is used by developers to manually input the latest research results, standard processes, and reference data into the knowledge base;

[0042] The question and answer data collection submodule is used to collect question and answer data between users and the system;

[0043] The question and answer data processing submodule is used to perform preprocessing operations including deduplication, filtering, cleaning, and labeling on the acquired question and answer data.

[0044] The system further includes a memory, a processor, and an electronic program stored in the memory and capable of running on the processor.

[0045] The present invention has the following beneficial effects:

[0046] Compared with the prior art, the system proposed by the technical solution of the present invention can collect user conversation data, clean and annotate the data, and after data cleaning, the system can fine-tune the original large model to make up for the problem of the large model not updating data in a timely manner, thereby improving the accuracy of the large model's answers. This mechanism ensures that the system can continuously learn and adapt to user needs and feedback, and keep its analysis and reasoning up-to-date and accurate;

[0047] Compared with the prior art, the system proposed in the technical solution of the present invention uses a large model to conduct in-depth mental health data analysis and reasoning. Through natural language processing technology and machine learning algorithms, the large model can accurately identify and evaluate the psychological state of adolescents, including but not limited to anxiety, depression, stress and social disorders. This analysis can not only provide real-time psychological state assessment, but also provide an important basis for subsequent mental health management services;

[0048] Compared with the prior art, the system proposed in the technical solution of the present invention realizes dynamic scheduling and management of adolescent mental health management services by setting up a task scheduling engine. The task scheduling engine can dynamically adjust the execution order and content of tasks according to the real-time needs of users and the analysis results of the large model, so that the system always provides the most suitable mental health services, including personalized psychological assessment, emergency intervention, long-term management and other services, which is conducive to the comprehensive and continuous attention and management of the mental health of adolescent users;

[0049] Compared with the prior art, the system proposed by the technical solution of the present invention can realize the centralized management and efficient access of mental health data, reduce the problems of data isolation and redundancy, and improve the efficiency of data management;

[0050] Compared with the prior art, the system proposed by the technical solution of the present invention continuously improves the analytical ability of the large model through self-learning and dynamic updating of the large model, and can provide more accurate and in-depth psychological state assessment. Continuous improvement ensures that the system can still maintain a high level of analysis and reasoning ability when facing new problems and new data;

[0051] Compared with the prior art, the advanced task scheduling engine set up in the system proposed in the technical solution of the present invention can dynamically adjust the provision of mental health services according to individual needs and real-time data, so that the system can flexibly respond to the different mental states and needs of users, provide personalized mental health management services, and improve the effectiveness of services and user experience;

[0052] Compared with the prior art, the system proposed by the technical solution of the present invention can not only receive the knowledge input by developers by building a knowledge base, but also expand and optimize in real time according to the question and answer data of users and the system. When the question and answer data accumulates to a certain amount, the big model learns the knowledge of these data, thereby continuously improving its professional ability and response accuracy.

[0053] Compared with the prior art, the system proposed by the technical solution of the present invention can provide high-quality mental health services including personalized psychological assessment, accurate emotion recognition and professional intervention suggestions by comprehensively utilizing the analytical capabilities of the large model and the resource support of the knowledge base;

[0054] Compared with the prior art, the system proposed in the technical solution of the present invention can continuously optimize the task scheduling strategy and model performance by collecting and analyzing user feedback and service data. This continuous improvement mechanism ensures that the system can continuously improve service quality and user satisfaction in long-term use. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is an interactive schematic diagram of the adolescent mental health management system based on knowledge base and AI of the present invention;

[0056] Figure 2 It is a schematic diagram of the large model self-learning process in the technical solution of the present invention. DETAILED DESCRIPTION

[0057] The present invention will be further described below in conjunction with specific implementation methods, wherein the accompanying drawings are only used for exemplary descriptions and represent only schematic diagrams rather than actual drawings, and should not be understood as limiting the present invention. In order to better illustrate the specific implementation methods of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product.

[0058] The present invention is based on artificial intelligence and big data technology. In the adolescent mental health system, it combines the analytical reasoning ability of the big model with the dynamic updating and optimization of the knowledge base to achieve more accurate services. The specific implementation process of the technical solution is as follows:

[0059] 1) Requirements analysis and task identification

[0060] Data collection: Collect data from adolescent user input, daily interactions, psychological assessments, and other data sources;

[0061] Big model analysis: Use big models to analyze the collected data and identify the mental health needs of adolescents, such as anxiety, depression, social disorders, etc. The big model analyzes the user's text input through natural language processing (NLP) technology to identify emotions and emotional states;

[0062] Using the emotion perception optimization model, EPOM combines LSTM and Attention mechanism, and makes full use of the emotion analysis ability of the large model, so that it has higher accuracy and efficiency in identifying user emotions and emotional states;

[0063] In the specific implementation process, the EPOM model first encodes the input text through the LSTM network to capture the contextual information and emotional changes in the text. Subsequently, the Attention mechanism performs weighted processing on the encoded information and dynamically focuses on the important parts of the text, thereby improving the accuracy of emotion recognition. Finally, through the sentiment analysis capabilities of the large model and the output results of the EPOM model, a comprehensive evaluation of the user's emotions and emotional state is performed. The LSTM network performs well in processing time series data and can capture the emotions and emotional changes in the user's input text. The Attention mechanism can dynamically focus on the key parts of the input text during the text analysis process, improving the model's ability to recognize emotions and emotional states. By combining the LSTM and Attention mechanisms and further optimizing them based on the sentiment analysis capabilities that the large model already has, the EPOM model can more accurately identify and analyze the emotions and emotional states of teenagers.

[0064] Knowledge base retrieval: In the process of analyzing needs, the big model can query the knowledge base to obtain information and resources related to the user’s current status, further enriching the results of the needs analysis.

[0065] 2) Task definition and priority setting

[0066] Based on the identified needs, define and prioritize the corresponding tasks:

[0067] Task definition: Each mental health need corresponds to one or more tasks, such as psychological assessment, emotional management, behavioral guidance, etc. When defining tasks, you can refer to the standard processes and suggestions in the knowledge base.

[0068] Priority setting: Set a priority for each task based on factors such as the urgency of the need and the scope of impact. The big model can determine the priority of tasks through historical data and predictive analysis.

[0069] 3) Dynamic task scheduling

[0070] Use AI technology to dynamically schedule tasks and adjust the execution order and content of tasks according to real-time conditions:

[0071] Real-time monitoring: Real-time monitoring of users’ psychological states and interactions to ensure the timeliness of task scheduling. The big model can provide real-time analysis results to help adjust task scheduling.

[0072] Dynamic adjustment: Based on real-time monitoring data and user feedback, the execution order and content of tasks are dynamically adjusted to ensure that the system always provides the most suitable mental health services.

[0073] 4) Intelligent task allocation

[0074] During task execution, AI technology is used to intelligently allocate tasks to optimize service quality:

[0075] Task allocation: Intelligently allocate tasks to appropriate service providers (such as counselors, psychologists, or automated service modules) based on task priority and user needs. The knowledge base can provide best practices and reference materials for task execution to help service providers better complete tasks.

[0076] 5) Task execution and monitoring

[0077] The task scheduling engine assigns tasks to corresponding execution units and monitors the entire process:

[0078] Task execution: Tasks are assigned to the appropriate mental health service providers to ensure they are executed as planned. During execution, the big model can analyze user feedback in real time and provide immediate adjustment suggestions.

[0079] Task monitoring: Monitor the progress and effect of the task throughout the process, and ensure that the task is completed as planned through a feedback mechanism. The data in the knowledge base can be used for comparison and reference to ensure the standardization and effectiveness of task execution.

[0080] 6) Result feedback and continuous optimization

[0081] After the task is completed, the system will evaluate the execution results and continuously optimize based on the evaluation results:

[0082] Result evaluation: Evaluate the performance of each task and collect user feedback and system data. The big model can analyze this data and provide improvement suggestions.

[0083] Continuous optimization: According to the evaluation results, adjust the task scheduling strategy and algorithm to continuously optimize the system's task scheduling and service content. The latest research and practical experience in the knowledge base can be used to update and optimize the system.

[0084] 7) There are several ways to dynamically update and optimize the knowledge base:

[0085] a. Internal input: Developers can manually input and update knowledge base content, including the latest research results, standard processes, and reference materials.

[0086] b. Question and answer data between users and the system: The system automatically collects question and answer data between users and the system, and updates the knowledge base in real time based on this data. For example, when a user asks a new question or gives feedback, the system will record and analyze this data and add the effective information to the knowledge base.

[0087] 8) Regular update of large models

[0088] When the question-and-answer data accumulates to a certain amount, the big model can learn the knowledge from these data and continuously strengthen its professional ability in adolescent mental health. Figure 2 , the specific steps are as follows:

[0089] The system collects question and answer data, and the data sources include: questions and answers entered externally and recorded in system logs

[0090] Data cleaning and labeling: When the amount of question and answer data reaches the upper limit, the collected question and answer data will be cleaned and labeled using similarity algorithms and keyword filtering to ensure the accuracy and validity of the data.

[0091] Model fine-tuning: Use cleaned and labeled data to regularly fine-tune large models to improve their analysis and reasoning capabilities.

[0092] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. Adolescent mental health management system based on knowledge base and AI, characterized by: The implementation process of the system includes the following steps: Collect data including input from adolescent users, daily interactions, psychological assessments, and other sources; Use big models to analyze the collected data to obtain the mental health needs of adolescents; Based on the results of the identification of mental health needs, define at least one task corresponding to the mental health needs and set the priority of the tasks; Use AI technology to dynamically schedule tasks, assign tasks to corresponding service providers, and adjust the execution order and content of tasks according to real-time conditions: The task scheduling engine allocates tasks to corresponding execution units and monitors the entire process. Obtain user feedback during the task execution process, analyze user feedback in real time through a large model, and provide immediate adjustment suggestions.

2. The adolescent mental health management system based on knowledge base and AI according to claim 1 is characterized by: The implementation process of the system also includes the following steps: Construct a youth mental health data knowledge base and dynamically update the knowledge base, wherein the knowledge base at least includes classified data on youth mental health needs, standard processes and recommended data for all tasks, best practices and reference data for all task executions, execution progress and effect data for all tasks, and the latest research and practical experience data on youth mental health.

3. The adolescent mental health management system based on knowledge base and AI according to claim 2 is characterized by: The knowledge base dynamic update methods include: Method 1: System developers manually input the latest research results, standard processes, and reference data into the knowledge base to update the knowledge base content; Method 2: Collect question and answer data between users and the system, and update the knowledge base in real time based on the obtained question and answer data.

4. The adolescent mental health management system based on knowledge base and AI according to claim 1 is characterized by: The implementation process of the system also includes the following steps: An upper limit value of the question-and-answer data of the user and the system is set. When the question-and-answer data accumulates to the upper limit value, the acquired question-and-answer data is learned through the big model to update the big model. The specific process includes the following steps: Collecting question and answer data, where the sources of the question and answer data include external input and system log records; Data cleaning and labeling: When the amount of question and answer data reaches the upper limit, the collected question and answer data will be preprocessed; Model fine-tuning, using the preprocessed question-answering data to fine-tune the large model.

5. The adolescent mental health management system based on knowledge base and AI according to claim 4 is characterized by: The implementation process of question-answer data preprocessing includes the following steps: The collected question and answer data is deduplicated through similarity algorithms; Set keywords and use them to filter and clean the collected question and answer data; Annotate the collected question-answering data.

6. The adolescent mental health management system based on knowledge base and AI according to claim 1 is characterized in that: It includes data collection module, large model module, knowledge base, task scheduling engine module, and user feedback acquisition module; The data collection module is used to collect data including input from adolescent users, daily interactions, psychological assessments and other sources; The large model module is used to identify the emotional and emotional state of the input user, obtain the mental health needs of the adolescent, define at least one corresponding task according to the mental health needs, set the priority of the task, dynamically schedule the task, and assign the task to the corresponding service provider: The task scheduling engine module is used to assign tasks to corresponding execution units and monitor the entire process; The user feedback acquisition module is used to obtain user feedback during the execution of the task, analyze the user feedback in real time through the big model, and provide instant adjustment suggestions.

7. The adolescent mental health management system based on knowledge base and AI according to claim 6 is characterized in that: The large model module includes a mental health needs analysis submodule, a task definition submodule, a task scheduling submodule, and a model update submodule. The mental health needs analysis submodule is used to analyze the collected data, analyze the text input of the teenagers through natural language processing technology, identify the emotions and emotional states of the input users, and obtain the mental health needs of the teenagers; The task definition submodule is used to define at least one task corresponding to the mental health need according to the identification result of the mental health need, and set the priority of the task; The task scheduling submodule uses AI technology to dynamically schedule tasks, assign tasks to corresponding service providers, and adjust the execution order and content of tasks according to real-time conditions: The model updating submodule is used to preprocess the collected question and answer data, learn from the preprocessed question and answer data, and fine-tune the large model according to the learning results.

8. The adolescent mental health management system based on knowledge base and AI according to claim 6 is characterized in that: The system also includes a knowledge base dynamic update module for updating the knowledge base content, and the knowledge base dynamic update module includes a data input submodule, a question and answer data collection submodule, and a question and answer data processing submodule; The data input submodule is used by developers to manually input the latest research results, standard processes, and reference data into the knowledge base; The question and answer data collection submodule is used to collect question and answer data between users and the system; The question and answer data processing submodule is used to perform preprocessing operations including deduplication, filtering, cleaning, and labeling on the acquired question and answer data.

9. The adolescent mental health management system based on knowledge base and AI according to claim 6 is characterized in that: The system also includes a memory, a processor, and an electronic program stored in the memory and capable of running on the processor, wherein the processor can implement the steps of the implementation process described in any one of claims 1-5 when running the electronic program.

Images