Student psychological data analysis and early warning system based on artificial intelligence

Abstract

The present invention relates to the field of health information processing technology, and specifically discloses an artificial intelligence-based student psychological data analysis and early warning system. The system sets a data set complexity determination module, a data slice determination module, a GPU compatible matching module, and a psychological data analysis and early warning module. The psychological health monitoring platform collects multi-source student psychological data sets, evaluates their comprehensive complexity, and determines the number of psychological data slices and their respective complexities based on this. The platform monitors the execution data of each GPU in real time, evaluates its initial execution quality indicators and compatible complexity. Subsequently, through data processing, the comprehensive complexity of the psychological data slices is matched with the compatible complexity of each GPU, and the allocation of psychological data slices to GPUs is completed. Finally, the platform uses this allocation relationship to analyze the students' psychological data, and ultimately achieves analysis and early warning of the students' psychological data.

Description

Technical Field

[0001] The present invention relates to the technical field of health information processing, and in particular to a student psychological data analysis and early warning system based on artificial intelligence. Background Art

[0002] The widespread application of artificial intelligence (AI) technology in various fields has yielded remarkable results, providing new insights and approaches to addressing student psychological data analysis and early warning. AI possesses powerful data processing, pattern recognition, and predictive analysis capabilities. It can deeply mine and analyze large amounts of student psychological data, identify potential signs of psychological problems, and issue timely warnings, providing strong support for mental health educators.

[0003] For example, the invention patent with announcement number CN118841178B announces a method and system for evaluating students' psychological literacy, involving the field of health data analysis technology. The entire evaluation system is divided into four parts, namely health data aggregation module, health assessment module, data storage and analysis module, and statistical adjustment module; its technical points are: collecting corresponding data according to the types of different students, which can more comprehensively understand and support the mental health needs of students of different age groups, and conduct comprehensive evaluation based on multiple factors of students' mental health reflected in the corresponding data and through analysis and calculation, which can more comprehensively capture the multi-dimensional information of students' mental health. When the psychological counseling mechanism is triggered subsequently, it helps to improve the efficiency and effectiveness of coping with students' mental health problems, and dynamically adjust the number of psychological teachers according to the proportion of the number of students who trigger the psychological counseling mechanism to the total number of students in the school, thereby realizing the effective allocation and utilization of resources.

[0004] For example, the invention patent with publication number CN119361142A discloses a mental health assessment and early warning analysis system based on health big data, including a data acquisition module, a monitoring module, an analysis and judgment module, an information management module and an early warning module. The monitoring module monitors the mental health status of all students in real time, and the analysis and judgment module analyzes the monitoring data to determine whether students have mental health problems, and timely warns when students have mental health problems through the early warning module.

[0005] Combining the above technical solutions, it is found that the monitoring data of students' mental health are more multi-source and the data volume is huge. The existing technical solutions for monitoring students' mental health directly use the acquired data without targeted processing, which easily brings unnecessary processing increments to the data processing equipment and reduces the processing speed of the equipment. At the same time, there is a problem of inaccurate psychological data analysis due to insufficient flexibility in equipment operation. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the present invention provides a student psychological data analysis and early warning system based on artificial intelligence, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: an artificial intelligence-based student psychological data analysis and early warning system, including a data set complexity determination module, which is used to extract multi-source psychological data sets through a mental health monitoring platform and comprehensively determine the comprehensive complexity of the multi-source psychological data sets; a data slice determination module, which is used to match the number of psychological data slices according to the comprehensive complexity of the multi-source psychological data sets, and finally extract the comprehensive complexity of each psychological data slice; a GPU compatible matching module, which is used to collect the execution data of each GPU of the mental health monitoring platform in real time, determine the initial execution quality index of each GPU, and match the compatible complexity of each GPU; a psychological data analysis and early warning module, which is used to process the comprehensive complexity of each psychological data slice with the compatible complexity of each GPU, complete the allocation of psychological data slices and GPUs, and finally the mental health monitoring platform analyzes and warns the student psychological data set.

[0008] As a further solution, a comprehensive complexity of a multi-source psychological dataset is comprehensively determined, specifically including: the multi-source psychological dataset includes each student's psychological dataset, the sub-complexity of each student's psychological dataset is determined and averaged, and finally the comprehensive complexity of the multi-source psychological dataset is obtained; a feature set is extracted from each student's psychological dataset based on the feature set, specifically including the total data volume of each student's psychological dataset, the number of data categories of each student's psychological dataset, the data time span of each student's psychological dataset, and the information entropy of each student's psychological dataset; the total data volume of each student's psychological dataset, the number of data categories of each student's psychological dataset, the time span of each student's psychological dataset, and the information entropy of each student's psychological dataset are weightedly aggregated and normalized in turn to obtain the sub-complexity of each student's psychological dataset.

[0009] As a further solution, the comprehensive complexity of each psychological data slice is extracted. The specific extraction process is: according to the number of psychological data slices, the multi-source psychological data set of each student is randomly divided to obtain each psychological data slice; the sub-complexity of each student's psychological data set contained in each psychological data slice is averaged to obtain the comprehensive complexity of each psychological data slice.

[0010] As a further solution, the initial execution quality indicators of each GPU are determined. The specific determination process is as follows: the execution data of each GPU, specifically including the number of floating-point operations of each GPU in the quality determination period, the instruction execution efficiency of each GPU in the quality determination period, the memory bandwidth of each GPU in the quality determination period, and the data throughput of each GPU in the quality determination period; the bounded number of floating-point operations, the bounded instruction execution efficiency, the bounded memory bandwidth, and the bounded data throughput are extracted from the mental health database; the degree of deviation between the number of floating-point operations of each GPU in the quality determination period and the bounded number of floating-point operations, the degree of deviation between the instruction execution efficiency of each GPU in the quality determination period and the bounded instruction execution efficiency, the degree of deviation between the memory bandwidth of each GPU in the quality determination period and the bounded memory bandwidth, and the degree of deviation between the data throughput of each GPU in the quality determination period and the bounded data throughput are weightedly aggregated in sequence to obtain the initial execution quality indicators of each GPU.

[0011] As a further solution, the allocation of psychological data slices and GPUs is completed. The specific analysis process is: the comprehensive complexity of each psychological data slice and the compatible complexity of each GPU are respectively subjected to difference processing to obtain the complexity deviation value between each GPU and each psychological data slice. The GPU corresponding to the minimum complexity deviation value between a certain psychological data slice is recorded as the designated GPU of the psychological data slice. The complexity deviation values between all GPUs and psychological data slices are traversed in turn to obtain the designated GPU of each psychological data slice, and the allocation of psychological data slices and GPUs is completed.

[0012] As a further solution, the student psychological data set is analyzed and early warning is carried out. The specific analysis process is: each psychological data slice is assigned to the corresponding designated GPU for data analysis, and the total processing delay time of the mental health monitoring platform is collected and compared with the predefined total processing delay definition time to obtain a comparison result. The mental health monitoring platform determines whether to adjust the processing efficiency of the GPU based on the comparison result. Finally, the mental health monitoring platform analyzes the student's psychological state and visualizes the analysis results as an early warning. The mental health monitoring platform outputs the early warning results to each receiving terminal to which it belongs.

[0013] As a further solution, the processing efficiency of the GPU is adjusted. The specific adjustment process is: the total processing delay time of the mental health monitoring platform is subtracted from the total processing delay definition time to obtain the delay time deviation value, and the clock frequency adjustment value is matched to adjust the processing efficiency of each GPU. At the same time, the execution quality real-time indicators of each GPU are monitored in real time and compared with the predefined execution quality threshold. If the execution quality real-time indicator of a GPU is less than the execution quality threshold, the core frequency adjustment value of the GPU is matched according to the difference between the execution quality real-time indicator of the GPU and the execution quality threshold, which is used to reduce the core frequency of the GPU.

[0014] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0015] (1) The present invention provides an artificial intelligence-based student psychological data analysis and early warning system. The psychological health monitoring platform collects multi-source student psychological data sets, evaluates their comprehensive complexity, and determines the number of psychological data slices and their respective complexities based on this. The platform monitors the execution data of each GPU in real time, evaluates its initial execution quality indicators and compatible complexity. Subsequently, through data processing, the comprehensive complexity of the psychological data slices is matched with the compatible complexity of each GPU, completing the allocation of psychological data slices to GPUs. Finally, the platform uses this allocation relationship to analyze the student's psychological data, ultimately achieving analysis and early warning of the student's psychological data.

[0016] (2) By extracting multi-source psychological data sets from each student and comprehensively determining the comprehensive complexity of the multi-source psychological data sets, the present invention can measure the complexity of the data as a whole, taking into account factors such as the total data volume, the number of data categories, the time span, and information entropy, helping the system understand the characteristics of the entire data set. This is conducive to the rational planning of data processing procedures, such as determining whether to adopt more efficient data processing algorithms or allocate more computing resources in advance based on the complexity, thereby laying the foundation for subsequent accurate analysis of student psychological data.

[0017] (3) The present invention can refine the data analysis by extracting the comprehensive complexity of each psychological data slice. After dividing the data set into slices, by calculating the complexity of each slice, the complexity differences between different parts of the data can be more accurately understood. In this way, when processing data, computing resources can be allocated in a targeted manner according to the complexity of each slice, and complex slices can be assigned to GPUs with higher performance to improve processing efficiency, while avoiding resource waste and ensuring more accurate and efficient analysis of students' psychological data. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of system module connections of the present invention.

[0020] Figure 2 This is a schematic diagram of the receiving terminals belonging to the mental health monitoring platform. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0022] See Figure 1 As shown, an embodiment of the present invention provides a technical solution: an artificial intelligence-based student psychological data analysis and early warning system, including a data set complexity determination module, a data slicing determination module, a GPU compatible matching module, and a psychological data analysis and early warning module.

[0023] An embodiment of the present invention provides a technical solution: an artificial intelligence-based student psychological data analysis and early warning system, which also includes: a mental health database for storing the number of floating-point operations, the efficiency of instruction execution, the bandwidth of video memory, the data throughput, and preset values of various factors.

[0024] The data set complexity determination module is connected to the data slicing determination module, the data slicing determination module is connected to the GPU compatible matching module, the GPU compatible matching module is connected to the psychological data analysis and early warning module, and the data set complexity determination module, the data slicing determination module, the GPU compatible matching module and the psychological data analysis and early warning module are all connected to the mental health database.

[0025] The dataset complexity determination module is used to extract multi-source psychological datasets through the mental health monitoring platform and comprehensively determine the comprehensive complexity of the multi-source psychological datasets.

[0026] The aforementioned mental health monitoring platform, built on big data and artificial intelligence technologies, targets the mental health needs of K-12 students. It integrates psychological assessment, crisis warning, and intervention support into an intelligent system (e.g., the Xiaozhi model). Its core functions include: integrating multiple sources of data, including online (student message interactions, mood logs, and "shudong" information) and offline (behavioral data and psychological assessment results), to comprehensively capture each student's mental state; using algorithms to identify psychological crisis signals such as anxiety and depression, triggering warnings based on pre-set rules and providing intervention recommendations; generating student-specific mental health reports, leveraging AI digital human technology (e.g., "Xiaozhi Classmate") to provide immediate psychological counseling, and providing data-based reference for school counseling offices. Compared to traditional models, this platform reduces stigma, improves intervention convenience, and enhances assessment accuracy, enabling dynamic monitoring and targeted intervention for students' mental health.

[0027] Specifically, the comprehensive complexity of the multi-source psychological dataset is comprehensively determined, including:

[0028] The multi-source psychological dataset includes the psychological datasets of each student. The sub-complexity of each student's psychological dataset is determined and averaged, and finally the comprehensive complexity of the multi-source psychological dataset is obtained.

[0029] A feature set is extracted based on each student psychological data set, specifically including the total data volume of each student psychological data set, the number of data categories of each student psychological data set, the data time span of each student psychological data set, and the information entropy of each student psychological data set.

[0030] The above information entropy can be specifically calculated using the existing Shannon entropy formula to measure the uncertainty of the data distribution of each student's psychological data set. The higher the entropy value, the greater the discreteness of the data and the higher the complexity.

[0031] The total data volume of each student's psychological data set, the number of data categories of each student's psychological data set, the time span of each student's psychological data set, and the information entropy of each student's psychological data set are weighted, aggregated, and normalized in turn to obtain the sub-complexity of each student's psychological data set. The specific analysis method is as follows:

[0032]

[0033] Where, is the subcomplexity of the i-th student psychological dataset, i is the number of each student psychological dataset, , M is the total amount of student psychological data sets, is the total data volume of the i-th student psychological data set, is the number of data categories in the i-th student psychology dataset, is the time span of the i-th student psychological dataset, is the information entropy of the i-th student psychological data set, is the weight factor corresponding to the total amount of data predefined in the mental health database, is the weight factor corresponding to the number of predefined data categories in the mental health database, is the weight factor corresponding to the predefined time span in the mental health database, It is the weight factor corresponding to the information entropy predefined in the mental health database.

[0034] It should be explained that the sub-complexity of each student psychological dataset mentioned above is represented by the individual complexity corresponding to each student psychological dataset, which is used to obtain the comprehensive complexity of the multi-source psychological dataset through mean processing.

[0035] It needs to be explained that the above-mentioned total data volume refers to the total number of records contained in a single student's psychological data set, reflecting the size of the data; the number of data categories refers to the number of different data types contained in the data set, such as structured assessment scores, semi-structured text feedback, unstructured audio and video records, etc.; the time span refers to the time range of data collection, such as the range from enrollment to the present time, reflecting the timeliness and continuity of the data.

[0036] The weight factors corresponding to the total data volume, the weight factors corresponding to the number of data categories, the weight factors corresponding to the time span, and the weight factors corresponding to the information entropy are all extracted from the mental health database, and the mapping relationship can be a one-to-one correspondence or a one-to-many relationship. For example, the total data volume, the number of data categories, the time span, and the information entropy are respectively mapped with the weight factors corresponding to the total data volume, the weight factors corresponding to the number of data categories, the weight factors corresponding to the time span, and the weight factors corresponding to the information entropy preset in the mental health database to form a mapping set. The real-time total data volume, the number of data categories, the time span, and the information entropy are brought into the mapping set to obtain the weight factors corresponding to the total data volume, the weight factors corresponding to the number of data categories, the weight factors corresponding to the time span, and the weight factors corresponding to the information entropy.

[0037] In this embodiment, the correlation between these parameters is considered through multivariate analysis of the total data volume, the number of data categories, the time span, and the information entropy. The larger the total data volume, the larger the data scale and the richer the data types usually covered (such as behavioral logs, text feedback, and audio and video records), which leads to an increase in information entropy, thereby greatly increasing the sub-complexity of the student psychological data set; when the time span is larger, long-term data (such as psychological files covering 3 years) may become more dispersed due to changes in the student's psychological state, resulting in an increase in the information entropy value, which has a negative impact on the sub-complexity of the student psychological data set.

[0038] The data slice determination module is used to match the number of psychological data slices according to the comprehensive complexity of multi-source psychological data sets, and finally extract the comprehensive complexity of each psychological data slice.

[0039] Furthermore, the number of psychological data slices is obtained by matching. The specific matching process is as follows:

[0040] The comprehensive complexity of the multi-source psychological data set is matched with the number of psychological data slices corresponding to each predefined comprehensive complexity interval, the interval to which the comprehensive complexity of the multi-source psychological data set belongs is determined, and the number of psychological data slices corresponding to the interval is obtained and used as the psychological data slices of the psychological data set.

[0041] Specifically, the comprehensive complexity of each psychological data slice is extracted. The specific extraction process is as follows:

[0042] According to the number of psychological data slices, the multi-source psychological data set of each student is randomly divided to obtain each psychological data slice.

[0043] The aforementioned psychological data slices specifically refer to sub-datasets formed by randomly partitioning a multi-source psychological dataset for multiple students. Each slice contains all the records of the original data for multiple students, which is used to distribute the processing load and match GPU resources.

[0044] Random partitioning avoids large deviations in data characteristics between slices, ensuring that the combined complexity of each slice is closer to the average of the entire dataset. This reduces drastic changes in slice complexity caused by time spans or localized concentrations of information entropy. Evenly distributed slices are more easily aligned with the GPU's compatible complexity, avoiding resource congestion caused by excessively complex slices. Furthermore, slice-level granularity allows for clearer resource scheduling, for example, allocating high-complexity slices to high-performance GPUs and low-complexity slices to standard GPUs, maximizing cluster efficiency.

[0045] The sub-complexities of each student's psychological data set contained in each psychological data slice are averaged to obtain the comprehensive complexity of each psychological data slice.

[0046] The GPU compatible matching module is used to collect the execution data of each GPU in the mental health monitoring platform in real time, determine the initial execution quality indicators of each GPU, and match the compatible complexity of each GPU.

[0047] Furthermore, the initial execution quality indicators of each GPU are determined. The specific determination process is as follows:

[0048] The execution data of each GPU specifically includes the number of floating-point operations of each GPU during the quality determination cycle, the instruction execution efficiency of each GPU during the quality determination cycle, the memory bandwidth of each GPU during the quality determination cycle, and the data throughput of each GPU during the quality determination cycle.

[0049] The execution data of each GPU can be extracted from the execution report of each GPU. The quality determination period is a period of time used to determine the execution quality of a GPU.

[0050] The bounded number of floating-point operations, bounded instruction execution efficiency, bounded video memory bandwidth, and bounded data throughput were extracted from the mental health database.

[0051] The initial execution quality indicators of each GPU are obtained by weighted aggregation of the deviation between the number of floating-point operations of each GPU during the quality determination period and the defined number of floating-point operations, the deviation between the instruction execution efficiency of each GPU during the quality determination period and the defined instruction execution efficiency, the deviation between the memory bandwidth of each GPU during the quality determination period and the defined memory bandwidth, and the deviation between the data throughput of each GPU during the quality determination period and the defined data throughput. The specific analysis method is as follows:

[0052]

[0053] Where, is the initial index of execution quality of the jth GPU, j is the number of each GPU, , D is the total number of GPUs, is the number of floating-point operations of the jth GPU in the quality determination cycle, Defines the number of floating-point operations, is the instruction execution efficiency of the jth GPU in the quality determination cycle, Define efficiency for instruction execution, is the memory bandwidth of the jth GPU during the quality determination period, Defines bandwidth for video memory, To define data throughput, is the data throughput of the jth GPU during the quality determination period, is the weight factor corresponding to the number of floating-point operations predefined in the mental health database, is the weight factor corresponding to the instruction execution efficiency predefined in the mental health database, is the weight factor corresponding to the video memory bandwidth predefined in the mental health database, The weight factor corresponding to the predefined data throughput in the mental health database.

[0054] The aforementioned initial execution quality indicators for each GPU measure GPU performance and functionality through a series of key parameters such as the number of floating-point operations and video memory bandwidth. These initial execution quality indicators directly affect the efficiency, accuracy, and stability of the GPU when executing tasks.

[0055] It should be explained that the above-mentioned instruction execution efficiency refers to the number of instructions that the GPU processor can execute per second. It is one of the important indicators for measuring the performance of the GPU processor, and the unit is millions of instructions per second.

[0056] It's important to explain that the floating-point operations (FLOPS) mentioned above refer to the total number of floating-point operations completed by the GPU during the quality determination cycle, reflecting its processing power for compute-intensive tasks (such as matrix multiplication and convolution). Instruction execution efficiency refers to how effectively the GPU executes instructions; a higher value means less instruction waiting time and less wasted resources. Memory bandwidth refers to the data transfer rate (GB / s) between the GPU's memory and cores, directly impacting the efficiency of large-scale data reads and writes. Data throughput refers to the number of tasks the GPU can handle per unit time. Higher data throughput indicates greater GPU processing power.

[0057] The weight factors corresponding to the number of floating-point operations, the weight factors corresponding to the instruction execution efficiency, the weight factors corresponding to the memory bandwidth, and the weight factors corresponding to the data throughput are all extracted from the mental health database, and the mapping relationship can be a one-to-one correspondence or a one-to-many relationship. For example, the number of floating-point operations, the instruction execution efficiency, the memory bandwidth, and the data throughput are respectively mapped with the weight factors corresponding to the number of floating-point operations, the weight factors corresponding to the instruction execution efficiency, the weight factors corresponding to the memory bandwidth, and the weight factors corresponding to the data throughput preset in the mental health database to form a mapping set. The real-time number of floating-point operations, the instruction execution efficiency, the memory bandwidth, and the data throughput are brought into the mapping set to obtain the weight factors corresponding to the number of floating-point operations, the weight factors corresponding to the instruction execution efficiency, the weight factors corresponding to the memory bandwidth, and the weight factors corresponding to the data throughput.

[0058] In this embodiment, the correlation between the number of floating-point operations, instruction execution efficiency, video memory bandwidth, and data throughput is considered through multivariate analysis. The number of floating-point operations is positively correlated with the video memory bandwidth. When the number of floating-point operations is high, frequent video memory access is required. Insufficient video memory bandwidth may cause idle computing units, forming a "memory wall" bottleneck, which has a negative impact on the initial indicators of GPU execution quality. When the instruction execution efficiency is high but the video memory bandwidth is low, data supply is not timely, and instructions cannot be executed due to waiting for data, thereby reducing GPU execution efficiency. Conversely, when the video memory bandwidth is high but the instruction execution efficiency is low, the data transmission speed is fast but the processing speed cannot keep up, and the GPU performance cannot be fully utilized, which affects execution quality. Data throughput is jointly affected by the number of floating-point operations, instruction execution efficiency, and video memory bandwidth. When all three indicators are at a high level, the GPU can quickly process a large number of tasks, increase data throughput, and improve execution quality. If any one of these indicators is low, data throughput will be limited, resulting in a decline in execution quality.

[0059] Specifically, the compatible complexity of each GPU is obtained by matching. The specific matching process is as follows:

[0060] The initial execution quality index of each GPU is matched with the compatible complexity corresponding to each predefined initial execution quality index interval, and the intervals to which the initial execution quality index of each GPU belongs are determined. The compatible complexity of each interval is obtained and assigned to the GPU corresponding to the initial execution quality index, and finally the compatible complexity of each GPU is obtained.

[0061] The psychological data analysis and early warning module is used to process the comprehensive complexity of each psychological data slice and the compatible complexity of each GPU, complete the allocation of psychological data slices and GPUs, and finally the mental health monitoring platform analyzes and warns students' psychological data sets.

[0062] Furthermore, the psychological data slices and GPU allocation are completed. The specific analysis process is as follows:

[0063] The comprehensive complexity of each psychological data slice and the compatible complexity of each GPU are respectively subjected to difference processing to obtain the complexity deviation value between each GPU and each psychological data slice. The GPU corresponding to the minimum complexity deviation value between a certain psychological data slice is recorded as the designated GPU of the psychological data slice. The complexity deviation values between all GPUs and psychological data slices are traversed in turn to obtain the designated GPU of each psychological data slice, thus completing the allocation of psychological data slices and GPUs.

[0064] It should be noted that the complexity deviation between each GPU and each psychological data slice can be zero, meaning that the combined complexity of a psychological data slice is equal to the GPU's compatible complexity. Furthermore, if the complexity deviation between each GPU and each psychological data slice is negative, meaning that the GPU's compatible complexity is less than the combined complexity of the psychological data slice, that GPU is discarded to ensure that all complexity deviations are natural numbers.

[0065] Psychological data slices are mapped one-to-one to specific GPUs, ensuring that high-complexity slices are allocated to high-performance GPUs and low-complexity slices are allocated to standard GPUs, significantly reducing resource waste or overload. A mapping strategy that minimizes complexity deviation reduces the processing time of individual slices on the GPU. Combined with real-time adjustment mechanisms (such as clock frequency adjustments), this ensures that overall processing latency remains below a threshold. Furthermore, if a GPU's real-time execution quality indicator falls below a threshold (e.g., core frequency anomalies), the system can quickly remap slices from that GPU to other devices, ensuring analysis continuity.

[0066] Specifically, the student psychological data set is analyzed and early warning is provided. The specific analysis process is as follows:

[0067] Each psychological data slice is assigned to the corresponding designated GPU for data analysis, and the total processing delay time of the mental health monitoring platform is collected, where the total processing delay time can be extracted from the operation log of the mental health monitoring platform and compared with the predefined processing delay limit total time to obtain a comparison result. The mental health monitoring platform determines whether to adjust the processing efficiency of the GPU based on the comparison result. Finally, the mental health monitoring platform analyzes the student's mental state and visualizes the analysis results for early warning. The mental health monitoring platform outputs the early warning result to each receiving terminal to which it belongs. Specifically, the mental health monitoring platform extracts the text data of each student's mental state data, and uses the built-in keyword retrieval algorithm of the mental health monitoring platform, such as the BM algorithm and the hash search algorithm, to perform keyword retrieval on the text data of each student's mental state data. For example, its keyword retrieval can be anxiety, depression, stress, shame, bullying, etc., and obtains the keyword retrieval number of each student, which is verified with the preset keyword retrieval limit number. If the keyword retrieval number of a student is greater than or equal to the keyword retrieval limit number, the student's mental state is warned, and the mental health monitoring platform outputs the early warning result to the student terminal, parent terminal, and teacher terminal to which the student belongs, such as Figure 2 As shown, Figure 2 This is a schematic diagram of the receiving terminals belonging to the mental health monitoring platform.

[0068] In one specific embodiment, student psychological state analysis can be performed by acquiring student interaction data from the Xiaozhi Mental Education Big Data Platform, behavioral data from the Anzhi Cloud School Big Data Model, and metadata from the Anzhi Cloud Platform's basic data. Multimodal big data acquisition and processing technologies (such as the Xiaozhi model) are used to process student psychological data, and personalized services are provided to students using AI-powered digital human technology (such as Xiaozhi). Professional-level psychological assessment systems (such as Xiaozhi Assessment) are used to assess student psychological data, and comprehensive analysis is conducted using multi-factor cross-models (such as interaction models, consumption models, moral education models, and academic affairs models). A personalized mental health report is generated for students, showcasing indicators such as emotional trends and social activity, and providing personalized recommendations (such as "Anxiety index increased this week; meditation is recommended"). Teachers can use a visual interface to view the overall psychological state distribution of their class (e.g., heat map), warning lists, and intervention records, and support exporting analysis reports. Parents can receive a desensitized summary of their child's psychological state (e.g., "My child's mood fluctuated significantly this week"), and use AI-powered digital human technology (such as Xiaozhi) to provide family education guidance.

[0069] Furthermore, the mental health monitoring platform determines whether to adjust the GPU processing efficiency based on the comparison results. The specific adjustment process is as follows:

[0070] If the total processing delay time of the mental health monitoring platform is greater than or equal to the total processing delay definition time, the comparison result will be recorded as the first comparison result; if the total processing delay time of the mental health monitoring platform is less than the total processing delay definition time, the comparison result will be recorded as the second comparison result.

[0071] If the comparison result shows the first comparison result, the mental health monitoring platform determines that the processing efficiency of the GPU needs to be adjusted. If the comparison result shows the second comparison result, the mental health monitoring platform determines that the processing efficiency of the GPU does not need to be adjusted.

[0072] Specifically, the processing efficiency of the GPU is adjusted. The specific adjustment process is as follows:

[0073] The total processing delay time of the mental health monitoring platform is subtracted from the total processing delay definition time to obtain the delay time deviation value, and the clock frequency adjustment value is matched to adjust the processing efficiency of each GPU. At the same time, the execution quality real-time indicators of each GPU are monitored in real time and compared with the predefined execution quality threshold. If the execution quality real-time indicator of a GPU is less than the execution quality threshold, the core frequency adjustment value of the GPU is matched according to the difference between the execution quality real-time indicator of the GPU and the execution quality threshold, which is used to reduce the core frequency of the GPU.

[0074] In one specific embodiment, adjusting the processing efficiency of each GPU specifically involves setting the initial clock frequency of each GPU to 500 MHz and the clock frequency adjustment value to 200 MHz, then adjusting the real-time clock frequency of each GPU to (500+200) MHz to increase the processing efficiency of each GPU, while simultaneously reducing the core frequency of the GPU specifically involves setting the core frequency adjustment value of the GPU to 50 MHz, then adjusting the core frequency of the GPU to (500+200-50) MHz to reduce the core frequency of the GPU.

[0075] In this embodiment, the clock frequency is adjusted by matching the delay duration deviation value. Increasing the clock frequency can shorten the instruction execution cycle, speed up the floating-point operation speed, and thus reduce the slice processing time, which can significantly reduce the processing delay and improve real-time performance. By reducing the core frequency of the abnormal GPU, the side effects caused by high-frequency operation can be reduced, the heat dissipation pressure can be effectively alleviated, and performance degradation or hardware damage due to overheating can be avoided. At the same time, for GPUs whose execution quality real-time indicators are lower than the threshold, reducing the core frequency can reduce the instruction execution error rate (such as floating-point operation errors) and improve the reliability of task processing.

[0076] The above content is merely an example and explanation of the structure of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the scope of protection of the present invention.

Claims

1. The student psychological data analysis and early warning system based on artificial intelligence is characterized by: include: The dataset complexity determination module is used to extract multi-source psychological datasets through the mental health monitoring platform and comprehensively determine the comprehensive complexity of the multi-source psychological datasets; The data slice determination module is used to match the number of psychological data slices according to the comprehensive complexity of the multi-source psychological data set, and finally extract the comprehensive complexity of each psychological data slice; The GPU compatibility matching module is used to collect the execution data of each GPU in the mental health monitoring platform in real time, determine the initial indicators of the execution quality of each GPU, and match them to obtain the compatible complexity of each GPU; The psychological data analysis and early warning module is used to process the comprehensive complexity of each psychological data slice and the compatible complexity of each GPU, complete the allocation of psychological data slices and GPUs, and finally the mental health monitoring platform analyzes and issues early warnings on the student psychological data set; The comprehensive determination of the comprehensive complexity of the multi-source psychological dataset specifically includes: The multi-source psychological dataset includes each student's psychological dataset, and the sub-complexity of each student's psychological dataset is determined and averaged to finally obtain the comprehensive complexity of the multi-source psychological dataset; Extracting feature sets based on each student's psychological data set, specifically including the total data volume of each student's psychological data set, the number of data categories of each student's psychological data set, the data time span of each student's psychological data set, and the information entropy of each student's psychological data set; The total data volume of each student's psychological data set, the number of data categories of each student's psychological data set, the time span of each student's psychological data set, and the information entropy of each student's psychological data set are weighted, aggregated, and normalized in turn to obtain the sub-complexity of each student's psychological data set; The specific determination process of determining the initial execution quality index of each GPU is as follows: The execution data of each GPU specifically includes the number of floating-point operations of each GPU during the quality determination period, the instruction execution efficiency of each GPU during the quality determination period, the memory bandwidth of each GPU during the quality determination period, and the data throughput of each GPU during the quality determination period; The bounded number of floating-point operations, bounded instruction execution efficiency, bounded video memory bandwidth, and bounded data throughput were extracted from the mental health database. The initial execution quality index of each GPU is obtained by weightedly aggregating the deviation between the number of floating-point operations of each GPU during the quality determination period and the defined number of floating-point operations, the deviation between the instruction execution efficiency of each GPU during the quality determination period and the defined instruction execution efficiency, the deviation between the memory bandwidth of each GPU during the quality determination period and the defined memory bandwidth, and the deviation between the data throughput of each GPU during the quality determination period and the defined data throughput. The matching process obtains the compatible complexity of each GPU. The specific matching process is as follows: The initial execution quality index of each GPU is matched with the compatible complexity corresponding to each predefined initial execution quality index interval, and the intervals to which the initial execution quality index of each GPU belongs are determined. The compatible complexity of each interval is obtained and assigned to the GPU corresponding to the initial execution quality index, and finally the compatible complexity of each GPU is obtained.

2. The student psychological data analysis and early warning system based on artificial intelligence according to claim 1 is characterized by: The matching obtains the number of psychological data slices, and the specific matching process is as follows: The comprehensive complexity of the multi-source psychological data set is matched with the number of psychological data slices corresponding to each predefined comprehensive complexity interval, the interval to which the comprehensive complexity of the multi-source psychological data set belongs is determined, and the number of psychological data slices corresponding to the interval is obtained and used as the psychological data slices of the psychological data set.

3. The student psychological data analysis and early warning system based on artificial intelligence according to claim 1 is characterized by: The comprehensive complexity of extracting each psychological data slice is specifically extracted as follows: According to the number of psychological data slices, the multi-source psychological data set of each student is randomly divided to obtain each psychological data slice; The sub-complexities of each student's psychological data set contained in each psychological data slice are averaged to obtain the comprehensive complexity of each psychological data slice.

4. The student psychological data analysis and early warning system based on artificial intelligence according to claim 1 is characterized by: The above mentioned psychological data slices and GPU allocation are analyzed in detail as follows: The comprehensive complexity of each psychological data slice and the compatible complexity of each GPU are respectively subjected to difference processing to obtain the complexity deviation value between each GPU and each psychological data slice. The GPU corresponding to the minimum complexity deviation value between a certain psychological data slice is recorded as the designated GPU of the psychological data slice. The complexity deviation values between all GPUs and psychological data slices are traversed in turn to obtain the designated GPU of each psychological data slice, thus completing the allocation of psychological data slices and GPUs.

5. The student psychological data analysis and early warning system based on artificial intelligence according to claim 1 is characterized by: The analysis and early warning of student psychological data sets are carried out in the following specific analysis process: Each psychological data slice is assigned to the corresponding designated GPU for data analysis, and the total processing delay time of the mental health monitoring platform is collected and compared with the predefined total processing delay time to obtain a comparison result. The mental health monitoring platform determines whether to adjust the processing efficiency of the GPU based on the comparison result. Finally, the mental health monitoring platform analyzes the student's mental state and visualizes the analysis results as an early warning. The mental health monitoring platform outputs the early warning results to each receiving terminal.

6. The student psychological data analysis and early warning system based on artificial intelligence according to claim 5 is characterized by: The mental health monitoring platform determines whether to adjust the GPU processing efficiency based on the comparison results. The specific adjustment process is as follows: If the total processing delay duration of the mental health monitoring platform is greater than or equal to the total processing delay definition duration, the comparison result is recorded as the first comparison result; if the total processing delay duration of the mental health monitoring platform is less than the total processing delay definition duration, the comparison result is recorded as the second comparison result; If the comparison result shows the first comparison result, the mental health monitoring platform determines that the processing efficiency of the GPU needs to be adjusted. If the comparison result shows the second comparison result, the mental health monitoring platform determines that the processing efficiency of the GPU does not need to be adjusted.

7. The student psychological data analysis and early warning system based on artificial intelligence according to claim 6 is characterized by: The processing efficiency of the GPU is adjusted, and the specific adjustment process is as follows: The total processing delay time of the mental health monitoring platform is subtracted from the total processing delay definition time to obtain the delay time deviation value, and the clock frequency adjustment value is matched to adjust the processing efficiency of each GPU. At the same time, the execution quality real-time indicators of each GPU are monitored in real time and compared with the predefined execution quality threshold. If the execution quality real-time indicator of a GPU is less than the execution quality threshold, the core frequency adjustment value of the GPU is matched according to the difference between the execution quality real-time indicator of the GPU and the execution quality threshold, which is used to reduce the core frequency of the GPU.

Images