Attention assessment and adaptive training system and method

Abstract

The invention discloses an attention assessment and adaptive training system and method, and the system comprises a data collection module which is used for obtaining the basic information and initial attention assessment data of a user; the attention evaluation and layering module is used for evaluating the attention level of the user and dividing user levels; the training task matching module is used for matching training tasks of corresponding difficulty levels of the user from a training question bank; the training effect evaluation module is used for identifying the performance data, performing automatic analysis on the performance data and generating an evaluation report at the same time; and the attention judgment module is used for judging whether the attention of the user is changed or not. By integrating multi-source data acquisition, intelligent layered evaluation and a dynamic adaptive training mechanism, a closed-loop system covering the whole attention evaluation and training process is constructed, and the long-term training effect is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of personal attention analysis, and in particular to an attention assessment and adaptive training system and method. Background Art

[0002] Currently, attention assessment and training technologies have important applications in education, medical rehabilitation, and personal development, but existing solutions have significant limitations. Traditional attention assessments primarily rely on paper-and-pencil tests or computerized psychometric tools (such as the Continuous Performance Task (CPT) and the Stroop test). These assessments rely on static task performance and fail to capture the user's physiological state (such as heart rate variability and eye movements) or dynamic attention changes in real time, resulting in biased assessments and difficulty guiding personalized training. Some high-end solutions use electroencephalogram (EEG) devices to monitor brain activity (such as the ratio of theta waves to beta waves). While these devices can provide insights into brain states, they are expensive, inconvenient to wear, and require specialized personnel to operate, making them difficult to implement for long-term training in everyday life. Virtual reality (VR) and augmented reality (AR) technologies can provide immersive training environments, but their stringent hardware requirements and high development complexity make them expensive and difficult to adopt by the general public. Mobile applications, while readily available, often offer training through superficial games or task-based design, lacking specialized depth and addressing deeper needs. Although existing adaptive training systems attempt to adjust difficulty based on user performance, their mechanisms are simple and crude, and they fail to comprehensively consider physiological signals, behavioral data, and individual differences, resulting in insufficient personalization of training and difficulty adapting to real-time changes in user abilities. In addition, existing solutions generally lack multi-dimensional effect feedback and visual analysis, making it difficult for users to intuitively understand their progress and weaknesses, affecting their continued participation. Issues with data security and operational process complexity are also common. For example, the interaction between terminal devices and the cloud lacks a standardized interface, or the user interface design is not adapted to different age groups, further limiting the popularity and practicality of the system. In summary, existing technologies have systemic defects in evaluation accuracy, personalized adaptation, user experience, and long-term application feasibility. A more efficient, accurate, and easily popularizable solution is urgently needed. Summary of the Invention

[0003] In view of this, the present invention proposes an attention assessment and adaptive training system and method, which can achieve high-precision attention assessment, personalized training path optimization, and user engagement improvement through multi-source data fusion and dynamic adaptive training mechanism. The present invention provides the following technical solutions:

[0004] An attention assessment and adaptive training system, comprising:

[0005] Data collection module, used to obtain basic information of users and their initial attention assessment data;

[0006] An attention assessment and stratification module, configured to assess the user's attention level and classify the user into different levels based on the initial attention assessment data by calling a preset assessment question bank and grading model;

[0007] A training task matching module is used to match the user with a training task of a corresponding difficulty level from a training question bank based on the attention level evaluation result and the user level;

[0008] The training effect evaluation module is used to identify the user's performance data on the training task, perform automatic analysis on it, and generate an evaluation report;

[0009] The attention judgment module is used to judge whether the user's attention has changed through the evaluation report. If it has not changed, the training at the current difficulty level will continue. If it has changed, the module will return to re-evaluate and match the training task.

[0010] Optionally, the attention assessment and stratification module specifically includes:

[0011] An evaluation data acquisition unit is used to collect the user's physiological signals, eye movement signals and task performance data in real time while the user is performing the evaluation task, and integrate them into evaluation data;

[0012] An analysis model unit, configured to assign different weights to the user's evaluation data according to the user's basic information, and obtain the user's attention score through an analysis model based on the evaluation data;

[0013] The attention stratification unit is used to set a data judgment threshold, and by comparing the evaluation data with the data judgment threshold, the user is assigned to three difficulty levels: low, medium, and high according to preset standards.

[0014] Optionally, the training task matching module specifically includes:

[0015] A radar chart generating unit, configured to generate a multi-dimensional ability radar chart based on the user's reaction speed, accuracy, fatigue index, and error rate, while defining a grade standard for each dimension and marking the user's current ability grade distribution based on the user's attention level assessment result;

[0016] A task combination strategy unit, used to match the difficulty of training tasks based on the user's attention level evaluation results;

[0017] A data collection unit is used to collect behavioral data and physiological data of the user when performing training tasks. The behavioral data includes reaction time, task accuracy and error type, and the physiological data includes heart rate data and eye movement data;

[0018] The difficulty adjustment unit is used to set the evaluation node, realize the dynamic evaluation of the user performance according to the behavioral data, and perform the difficulty upgrade or difficulty reduction operation according to the dynamic evaluation result, wherein the difficulty upgrade operation only changes one training task parameter each time, and increases it in multiple steps. During the difficulty reduction operation, the original difficulty training task is retained, and simple tasks are inserted in multiple times to realize the step-by-step reduction of the training task difficulty.

[0019] Optionally, the training effect evaluation module specifically includes:

[0020] a data processing unit, configured to identify the user's behavioral data and physiological data for the training task, and perform outlier processing and normalization processing to obtain preprocessed data;

[0021] An evaluation model analysis unit, configured to calculate scores of multi-dimensional indicators based on the pre-processed data, and calculate an overall score based on preset weight parameters of the indicators of each dimension for different users;

[0022] The report generation unit is used to summarize the current user's multi-dimensional indicator scoring data, overall scoring data and performance data of historical training tasks, and perform performance analysis based on the summarized multiple data. At the same time, it provides improvement suggestions for the user and generates an evaluation report for the current user based on the summarized multiple data, performance analysis and improvement suggestions.

[0023] Optionally, the attention judgment module specifically includes:

[0024] The data analysis and attention detection unit is used to determine the user's initial attention baseline based on the user's initial attention assessment data and obtain the user's real-time training task performance data to calculate the user's attention change in different dimensional indicators;

[0025] A threshold determination unit, configured to define a threshold for improvement or deterioration of attention, and to determine changes in user attention by comparing the magnitude of the attention change with the threshold;

[0026] The dynamic adjustment strategy unit is used to dynamically adjust the user's training task based on the judgment result of the user's attention change. If the judgment result of the attention change is that there is no significant change, the training task of the user's current difficulty level will continue to be maintained. If the judgment result of the attention change is that there is a significant change, the re-evaluation process will be triggered, and the difficulty level of the training task will be updated through the attention baseline obtained by the re-evaluation of the user.

[0027] Optionally, a prediction module is also included, which is used to predict the user's future performance trend based on the user's current performance data using a pre-trained prediction model.

[0028] The present invention further discloses an attention assessment and adaptive training method, which is applied to the above-mentioned system, and the method comprises:

[0029] The data collection module obtains the user's basic information and the user's initial attention assessment data;

[0030] The attention assessment and stratification module calls a preset assessment question bank and grading model based on the initial attention assessment data to assess the user's attention level and divide the user into tiers;

[0031] The training task matching module matches the user with a training task of a corresponding difficulty level from the training question bank according to the attention level evaluation result and the user level;

[0032] The training effect evaluation module identifies the user's performance data on the training task, performs automatic analysis on it, and generates an evaluation report;

[0033] The attention judgment module determines whether the user's attention has changed through the evaluation report. If not, the training at the current difficulty level will continue. If there is a change, the module will return to re-evaluate and match the training task.

[0034] The present invention further discloses a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the functions of the above system are realized.

[0035] The present invention further discloses an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor realizes the functions of the above-mentioned system when executing the program.

[0036] The present invention further discloses a computer program product, comprising a computer program, characterized in that the computer program implements the functions of the above system when executed by a processor.

[0037] The technical solution of the present invention integrates multi-source data collection, intelligent hierarchical assessment, and a dynamic adaptive training mechanism to construct a closed-loop system covering the entire process of attention assessment and training. The data acquisition module simultaneously acquires the user's physiological signals, behavioral performance, and basic information, and combines dynamic weight allocation to achieve multi-dimensional data fusion, significantly improving the accuracy and real-time performance of attention assessment, and solving the one-sidedness problem caused by the traditional method's reliance on a single indicator. At the same time, the attention assessment and stratification module divides the ability level into three levels: low, medium, and high, using a preset hierarchical model. It is linked with the training task matching module. Based on the multi-dimensional ability distribution generated by radar charts, combined with a step-by-step difficulty adjustment, it ensures that the training difficulty is dynamically matched to the user's real-time ability, avoiding the lack of personalization caused by static rules in existing technologies. The training effect evaluation module uses standardized data processing and a weighted scoring model, combined with visual reports to compare historical data and generate improvement suggestions, quantifying training results and enhancing user participation motivation. The attention judgment module compares the initial baseline with real-time data, triggering reassessment and updating the training difficulty in specific circumstances, forming a continuously optimized training closed loop, significantly improving long-term training effectiveness. This solves the systemic shortcomings of existing technologies in terms of assessment accuracy, dynamic adaptability, user experience, and popularity. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] For purposes of illustration and not limitation, the present invention will now be described with reference to embodiments thereof and the accompanying drawings, in which:

[0039] Figure 1 is a schematic structural diagram of an attention assessment and adaptive training system in an embodiment of the present invention;

[0040] Figure 2 is a flowchart of attention assessment and adaptive training in an embodiment of the present invention;

[0041] Figure 3 Schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of this application.

[0043] It should be noted that, in the absence of conflict, the embodiments of the present application and the features thereof can be combined with each other. The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0044] refer to Figure 1 , this embodiment discloses an attention assessment and adaptive training system, including:

[0045] The data collection module 11 is used to obtain the user's basic information and the user's initial attention evaluation data.

[0046] The data acquisition module is the core input unit of the system, which is used to realize the collection of basic user information, initial attention assessment data collection, data transmission and storage, and dynamic adaptation mechanism. In this embodiment, a hardware and software interface is designed to guide users to enter basic information through a graphical user interface (GUI), including fields such as age, gender, educational background, and previous training history. The interface uses interactive components such as drop-down menus and radio buttons to ensure the standardization of data formats (for example, only positive integers are allowed for age, and gender is "male / female / other" options). The input data is verified by real-time verification logic. For example, the age field uses regular expressions to filter non-numeric inputs, and mandatory items (such as age) are set with trigger verification prompts. After verification, the data is stored in a structured format (such as JSON) to a local or cloud database and is encrypted with AES-256 to protect privacy. This system integrates data acquisition interfaces for multi-source sensors to realize the collection of physiological data and behavioral data of users. Specifically, a wireless heart rate sensor (such as a photoplethysmography sensor, PPG) monitors the user's heart rate variability (HRV) in real time and records it as heart rate data. An eye tracking device records pupil coordinates, blink frequency, and gaze trajectory to form eye movement data, and stores the heart rate and eye movement data as physiological data. When the user performs the initial assessment task, a task response recorder simultaneously collects reaction time, task accuracy, error type, and task completion time to form initial behavioral data.

[0047] Physiological data is further fused and preprocessed. Specifically, heart rate data is filtered and features extracted. Kalman filtering is used to remove noise from eye movement data, extracting the gaze stability index. The Z-score method is used to filter extreme data. A sliding window algorithm is used to detect data mutation points. Then, eye movement data and heart rate data are matched by timestamp alignment. Hardware trigger signals are then used to trigger and match the matched eye movement and heart rate data with behavioral data, ultimately forming the user's initial attention assessment data.

[0048] The attention assessment and stratification module 12 is used to call a preset assessment question bank and grading model to assess the user's attention level based on the initial attention assessment data and divide the user into tiers.

[0049] The attention assessment and stratification module specifically includes an assessment data acquisition unit, an analysis model unit, and an attention stratification unit. After obtaining the initial attention assessment data generated by the data acquisition module, the module calls the assessment questions suitable for the user in the preset assessment question bank based on the data.

[0050] Furthermore, while the user is performing the assessment task, the assessment data acquisition unit collects the user's physiological signals, eye movement signals, and task performance data in real time and fuses them into assessment data. Specifically, the user's real-time heart rate signal is collected and timestamped to form heart rate data; pupil coordinates, blink frequency, and gaze point trajectory are collected and recorded, and the gaze stability index and the number of microsaccades in the eye movement trajectory per unit time are calculated to form eye movement data; while the user is performing the assessment task, the following task performance data is recorded in real time: reaction time: the response time from the appearance of the task prompt to the user's operation; task accuracy: the accuracy rate is calculated by automatically comparing the user input with a preset answer library; error type distribution: distraction errors and operational errors are classified and counted. The acquired heart rate data, eye movement data, and task performance data are fused into assessment data.

[0051] After acquiring the evaluation data, the analysis model unit assigns different weights to the user's evaluation data based on the user's basic information, and analyzes the user's attention level based on the evaluation data. Specifically, the user's age data is identified based on the user's basic information. Users under 18 are defined as child users, and users 18 and older are defined as adult users. Different weights are then assigned to the different evaluation data for adult and child users. For example, the weights for child users are set as follows: heart rate data is weighted 0.6, eye movement data is weighted 0.2, and task performance data is weighted 0.2; while the weights for adult users are set as follows: heart rate data is weighted 0.3, eye movement data is weighted 0.5, and task performance data is weighted 0.2. By assigning different weights, the focus is placed on heart rate performance and eye movement stability for child users, while task performance is emphasized for adult users. Furthermore, further modifications can be made based on the user's gender; for female users, the weighting for eye movement is increased when assigning weights. The acquired evaluation data is normalized, and the weighted, normalized evaluation data is input into a pre-set analysis model to output a user's attention level score.

[0052] Furthermore, a data judgment threshold is set through the attention stratification unit, and by comparing the evaluation data with the data judgment threshold, the user is assigned to three difficulty levels: low, medium, and high according to preset standards. For example, the threshold for the low difficulty level is 60 points, and the threshold for the medium difficulty level is 85 points. At the same time, additional conditions can be set based on the judgment threshold. For example, when judging a user as a low difficulty level, the reaction time fluctuation coefficient must also be greater than 0.3; when judging a user as a medium difficulty level, the task accuracy rate must also meet ≥75% and the stability index must be ≥0.8.

[0053] The training task matching module 13 is used to match the user with a training task of a corresponding difficulty level from the training question bank based on the attention level assessment result and the user level. The training task matching module specifically includes a radar chart generation unit, a task combination strategy unit, a data collection unit, and a difficulty adjustment unit.

[0054] The radar chart generation unit is used to generate a multi-dimensional ability radar chart based on the user's reaction speed, accuracy, fatigue index and error rate, and at the same time define the grade standard for each dimension, and mark the user's current ability level distribution according to the user's attention level evaluation result. Specifically, the standardized processing results of the training effect evaluation module are called, and the hierarchical label (low / medium / high) of the attention evaluation module is read at the same time, and the original data is mapped to a unified dimension of 0-100 points. The grade standard definition adopts a hierarchical dynamic threshold mechanism:

[0055] Reaction speed: divided into 5 levels (level 1: ≤60 points, level 2: 61-70 points, level 3: 71-80 points, level 4: 81-90 points, level 5: ≥91 points);

[0056] Accuracy: The 5-level standard is based on the task type. For example, an accuracy rate of 80% for basic tasks is level 4, and an accuracy rate of 90% for advanced tasks is level 5.

[0057] Fatigue index: graded by reaction time fluctuation coefficient, with a fluctuation coefficient of ≤0.15 being grade 5 (high stability) and ≥0.4 being grade 1 (significant fluctuation);

[0058] Error rate: Divided by error type distribution. For example, if distraction errors account for ≤10%, it is level 5; if >30%, it is level 1.

[0059] The task combination strategy unit is used to match the difficulty of training tasks based on the user's attention level assessment results. Specifically, for users with a low difficulty level, the main task is a basic task, accounting for 60%, supplemented by an anti-interference ability consolidation task of 20%, and an overlapping cross-dimensional task of 20%; for users with a medium difficulty level, the main task adopts dual-task switching (for example, processing visual and auditory instructions at the same time), accounting for 60%, consolidation tasks focus on the dominant dimension (for example, if the user's accuracy rate is ≥85%, time pressure questions are added), and cross-dimensional tasks strengthen weaknesses (for example, if the error rate is >25%, distraction interference training is inserted); for users with a high difficulty level, the main task is a complex task, accounting for 60%, consolidation tasks are high-precision maintenance training, and cross-dimensional tasks introduce multimodal interference.

[0060] After determining the difficulty of the training task matched to the user, the user performs the corresponding training task. The data collection unit collects behavioral and physiological data from the user during the training task. Behavioral data includes reaction time, task accuracy, and error types, while physiological data includes heart rate and eye movement data. Furthermore, the difficulty adjustment unit sets evaluation nodes and dynamically evaluates user performance based on the behavioral data. Difficulty escalation and reduction operations are performed based on the dynamic evaluation results. The difficulty escalation operation changes only one training task parameter at a time, increasing the difficulty in multiple steps. The difficulty reduction operation retains the original difficulty training task and inserts simpler tasks in multiple steps to achieve a step-by-step reduction in the difficulty of the training task. Specifically, fixed evaluation nodes and dynamic trigger nodes are set. The fixed evaluation node triggers an evaluation every three consecutive training tasks, while the dynamic trigger node triggers an evaluation immediately upon detecting abnormal data. After the evaluation is completed, the difficulty escalation or reduction operation is executed based on the evaluation results. Difficulty escalation requires the following conditions: accuracy for three consecutive tasks ≥ 85%; reaction time fluctuation coefficient (standard deviation / mean) < 0.2; and distraction error percentage < 10%. Difficulty downgrade requires any of the following conditions: accuracy of two consecutive tasks is less than 60%; reaction time increases suddenly (current average > historical average + 0.5 seconds); eye movement trajectory drift rate is > 15px / s and lasts for > 3 seconds. Regarding the specific difficulty adjustment strategy, the adjustment strategy for difficulty upgrade is a single parameter gradual adjustment:

[0061] Only one training task parameter is changed at a time, and the upgrade is carried out in three stages:

[0062] Phase 1: Only the number of distractors was increased, while the reaction time limit remained unchanged, and the training task lasted for 3 times.

[0063] Phase 2: Shorten the reaction time limit and maintain 4 distractors for 3 times.

[0064] Phase 3: Increase the number of distractions to 5 and shorten the time to 1.5 seconds to complete the final difficulty transition.

[0065] Adaptive threshold control: If the accuracy of the task drops by more than 15% after the upgrade, the "rollback mechanism" will be triggered, undoing the current stage adjustment and restoring the previous difficulty parameters.

[0066] The adjustment strategy for difficulty upgrade is parameter retention and insertion strategy:

[0067] Basic tasks are retained: 80% of the original difficulty level is maintained.

[0068] Simple task insertion: Insert a 20% "buffer task" (reduce distractions to 2, and relax the reaction time limit to 2.5 seconds), and gradually call back in three stages:

[0069] Phase 1: Insert 20% of simple tasks for 3 times;

[0070] Phase 2: Reduce to 15% simple tasks and reduce the number of distractors to 4;

[0071] Phase 3: Completely remove simple tasks, reduce the number of distractions to 3, and end the downgrade process.

[0072] If the user's eye movement data continues to show distraction, "Assistive Prompts" will be automatically turned on and the reaction time buffer will be extended.

[0073] After the difficulty adjustment strategy is completed, the adjusted parameters are passed to the task combination strategy unit via the API interface to generate a new task combination. For example, after upgrading, the difficulty of the "main task" is increased to medium, and the proportion of cross-dimensional tasks is increased to 30%. Real-time synchronization of radar chart data, for example, if the user's "reaction speed" dimension reaches level 4, the weight of the training task in this dimension is reduced (from 40% to 30%), and the proportion of "anti-interference ability" tasks is increased.

[0074] The training effect evaluation module 14 is used to identify the user's performance data on the training task, automatically analyze it, and generate an evaluation report. The training effect evaluation module specifically includes a data processing unit, an evaluation model analysis unit, and a report generation unit.

[0075] The data processing unit is used to identify the user's behavioral data and physiological data for the training task, and perform outlier processing and normalization processing to obtain pre-processed data;

[0076] The evaluation model analysis unit is used to calculate the scores of the multi-dimensional indicators based on the pre-processed data, and calculate the overall score based on the preset weight parameters of the dimensional indicators for different users;

[0077] The report generation unit is used to summarize the current user's multi-dimensional indicator scoring data, overall scoring data and performance data of historical training tasks, and perform performance analysis based on the summarized multiple data, while giving improvement suggestions for the user, and generating an evaluation report for the current user based on the summarized multiple data, performance analysis and improvement suggestions.

[0078] The attention determination module 15 is used to determine whether the user's attention has changed based on the evaluation report. If not, training at the current difficulty level continues. If so, re-evaluation and matching of training tasks are performed. The attention determination module specifically includes a data analysis and attention detection unit, a threshold determination unit, and a dynamic adjustment strategy unit.

[0079] Specifically, the data analysis and attention detection unit dynamically monitors the change in user attention through multi-dimensional data fusion and time series analysis. First, data collection is performed, and the multi-dimensional data of the initial evaluation phase is called. The initial attention baseline value is calculated by weighted average. This embodiment provides the calculation formula of the baseline value: Among them, F i is the standardized value of the i-th dimension, weight W i Dynamically allocate attention based on user level. Then, obtain user performance data on real-time training tasks. Finally, calculate the magnitude of change, including dimension-level changes. First, calculate the standardized difference for each metric, then use weighted summation to generate the overall magnitude of attention change. Use a moving average to filter short-term fluctuations and identify long-term trends.

[0080] After calculating the magnitude of the attention change, the threshold determination unit compares the preset dynamic threshold with the calculated magnitude of the change, ultimately determining whether the user's attention has changed significantly or not. Furthermore, based on the determination of the attention change, the dynamic adjustment strategy unit dynamically adjusts the user's training task based on the determination of the user's attention change. If the determination of the attention change is that there has been no significant change, the training task at the user's current difficulty level is maintained. If the determination of the attention change is that there has been a significant change, a reassessment process is triggered, and the difficulty level of the training task is updated based on the attention baseline derived from the reassessment of the user.

[0081] Prediction module 16 is used to predict a user's future performance trends based on their current performance data using a pre-trained prediction model. Specifically, the prediction model is developed using advanced technologies such as deep learning. This model can predict future performance trends based on a user's current performance. Specifically, it analyzes the user's most recent seven training sessions and predicts the reaction speed and accuracy of the next three training sessions. It also analyzes the user's most recent 28 training sessions to identify the overall trend of ability changes and generate a performance development curve for the user over the next month. It also integrates short-term and long-term prediction results and dynamically adjusts the weights of short-term and long-term predictions based on the user's current data volume. The combined prediction results include both recent performance and long-term trends. Finally, the prediction results are generated: an immediate prediction of the user's performance over the next three training sessions, which is used to dynamically adjust task difficulty; a mid-term prediction of the user's ability development trend over the next week, which helps formulate short-term training plans; and a long-term prediction of the user's ability changes over the next month, which is used to plan long-term training goals.

[0082] In summary, the system accurately quantifies attention levels by integrating physiological signals, behavioral data, and basic user information, combined with dynamic weight allocation and machine learning models; secondly, it adopts a step-by-step difficulty adjustment strategy to effectively improve the degree of training personalization, enhance user adaptability, and avoid frustration caused by sudden changes in difficulty; at the same time, it visualizes ability distribution through radar charts, significantly improves user engagement, and supports long-term training planning; in addition, the system simplifies the operating process, reduces hardware dependence, ensures data security and timeliness of training content, and ultimately forms a closed-loop optimization of "evaluation-training-feedback-adjustment", which solves the core problems of existing technologies such as one-sided evaluation, insufficient personalization, poor user experience, and low popularity.

[0083] At the same time, through a closed-loop mechanism combining multi-source data fusion, dynamic evaluation, and personalized training, this system addresses existing issues such as insufficient evaluation accuracy, low personalization, complex operation, and unpredictable long-term effects. The system first uses a multi-dimensional data acquisition module to acquire the user's physiological signals and behavioral data in real time. This data is dynamically weighted based on the user's basic information. A machine learning model then performs multi-dimensional analysis, generating a performance radar chart and stratifying attention levels. Based on this, the personalized training matching module employs a task combination strategy based on the radar chart's stratification results and monitors the user's task performance data in real time. This allows for smooth adjustments in training difficulty through step-by-step upgrades or soft-landing downgrades. This system further integrates a prediction model to generate short-term predictions based on the user's recent training data and a one-month performance trend curve based on long-term data. This allows for dynamic adjustments to training paths and long-term goal planning. Furthermore, an online learning mechanism automatically updates model parameters when prediction errors exceed a threshold. Finally, a feedback report presents intuitive multi-dimensional scores, historical comparisons, and improvement suggestions, forming a fully intelligent closed-loop system from evaluation to training, significantly improving evaluation accuracy, training efficiency, and user engagement. Compared to existing technologies, this application achieves higher levels of personalization and accuracy in both evaluation and training.

[0084] refer to Figure 2 This embodiment further discloses an attention assessment and adaptive training method, which uses the above-mentioned attention assessment and adaptive training system, and includes:

[0085] The data collection module obtains the user's basic information and the user's initial evaluation attention data;

[0086] The attention assessment and stratification module calls a preset assessment question bank and grading model based on the initial assessment attention data to assess the user's attention level and divide the user into tiers;

[0087] The training task matching module matches the user with a training task of a corresponding difficulty level from the training question bank according to the attention level evaluation result and the user level;

[0088] The training effect evaluation module identifies the user's performance data on the training task, performs automatic analysis on it, and generates an evaluation report;

[0089] The attention judgment module determines whether the user's attention has changed through the evaluation report. If not, the training at the current difficulty level will continue. If there is a change, the module will return to re-evaluate and match the training task.

[0090] Figure 3 This is a schematic diagram of the physical structure of an electronic device of the attention assessment and adaptive training system disclosed in an embodiment of the present invention. Figure 3As shown, the electronic device 50 includes: a processor 501 (processor), a memory 502 (memory) and a bus 503;

[0091] The processor 501 and the memory 502 communicate with each other via the bus 503 ; the processor 501 is used to call program instructions in the memory 502 to execute the methods provided by the above-mentioned method implementation methods.

[0092] This embodiment provides a non-transitory computer-readable storage medium, which stores computer instructions. The computer instructions enable a computer to execute the methods provided by the above-mentioned method embodiments.

[0093] Those skilled in the art will understand that all or part of the steps for implementing the above-mentioned method implementation method can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method implementation method; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk, etc. Various storage media that can store program codes.

[0094] The device embodiments described above are merely illustrative. Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of these modules may be selected based on actual needs to achieve the objectives of this embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0095] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or certain parts of the embodiment.

[0096] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. An attention assessment and adaptive training system, characterized in that: include: Data collection module, used to obtain basic information of users and their initial attention assessment data; An attention assessment and stratification module, configured to assess the user's attention level and classify the user into different levels based on the initial attention assessment data by calling a preset assessment question bank and grading model; A training task matching module is used to match the user with a training task of a corresponding difficulty level from a training question bank based on the attention level evaluation result and the user level; The training effect evaluation module is used to identify the user's performance data on the training task, perform automatic analysis on it, and generate an evaluation report; The attention judgment module is used to judge whether the user's attention has changed through the evaluation report. If it has not changed, the training at the current difficulty level will continue. If it has changed, the module will return to re-evaluate and match the training task.

2. The attention assessment and adaptive training system according to claim 1, characterized in that The attention assessment and stratification module specifically includes: An evaluation data acquisition unit is used to collect the user's physiological signals, eye movement signals and task performance data in real time while the user is performing the evaluation task, and integrate them into evaluation data; An analysis model unit, configured to assign different weights to the user's evaluation data according to the user's basic information, and obtain the user's attention score through an analysis model based on the evaluation data; The attention stratification unit is used to set a data judgment threshold, and by comparing the evaluation data with the data judgment threshold, the user is assigned to three difficulty levels: low, medium, and high according to preset standards.

3. The attention assessment and adaptive training system according to claim 1, characterized in that The training task matching module specifically includes: A radar chart generating unit, configured to generate a multi-dimensional ability radar chart based on the user's reaction speed, accuracy, fatigue index, and error rate, while defining a grade standard for each dimension and marking the user's current ability grade distribution based on the user's attention level assessment result; A task combination strategy unit, used to match the difficulty of training tasks based on the user's attention level evaluation results; A data collection unit is used to collect behavioral data and physiological data of the user when performing training tasks. The behavioral data includes reaction time, task accuracy and error type, and the physiological data includes heart rate data and eye movement data; The difficulty adjustment unit is used to set the evaluation node, realize the dynamic evaluation of the user performance according to the behavioral data, and perform the difficulty upgrade or difficulty reduction operation according to the dynamic evaluation result, wherein the difficulty upgrade operation only changes one training task parameter each time, and increases it in multiple steps. During the difficulty reduction operation, the original difficulty training task is retained, and simple tasks are inserted in multiple times to realize the step-by-step reduction of the training task difficulty.

4. The attention assessment and adaptive training system according to claim 1, characterized in that The training effect evaluation module specifically includes: a data processing unit, configured to identify the user's behavioral data and physiological data for the training task, and perform outlier processing and normalization processing to obtain preprocessed data; An evaluation model analysis unit, configured to calculate scores of multi-dimensional indicators based on the pre-processed data, and calculate an overall score based on preset weight parameters of the indicators of each dimension for different users; The report generation unit is used to summarize the current user's multi-dimensional indicator scoring data, overall scoring data and performance data of historical training tasks, and perform performance analysis based on the summarized multiple data. At the same time, it provides improvement suggestions for the user and generates an evaluation report for the current user based on the summarized multiple data, performance analysis and improvement suggestions.

5. The attention assessment and adaptive training system according to claim 1, characterized in that The attention judgment module specifically includes: The data analysis and attention detection unit is used to determine the user's initial attention baseline based on the user's initial attention assessment data and obtain the user's performance data of real-time training tasks to calculate the degree of change in the user's attention in different dimensional indicators; A threshold determination unit, configured to define a threshold for improvement or deterioration of attention, and to determine changes in user attention by comparing the magnitude of the attention change with the threshold; The dynamic adjustment strategy unit is used to dynamically adjust the user's training task based on the judgment result of the user's attention change. If the judgment result of the attention change is that there is no significant change, the training task of the user's current difficulty level will continue to be maintained. If the judgment result of the attention change is that there is a significant change, the re-evaluation process will be triggered, and the difficulty level of the training task will be updated through the attention baseline obtained by the re-evaluation of the user.

6. The attention assessment and adaptive training system according to claim 1, characterized in that The system also includes a prediction module for predicting the user's future performance trend based on the user's current performance data through a pre-trained prediction model.

7. A method for attention assessment and adaptive training, applied to the system according to any one of claims 1 to 6, characterized in that: The method includes: The data collection module obtains the user's basic information and the user's initial attention assessment data; The attention assessment and stratification module calls a preset assessment question bank and grading model based on the initial attention assessment data to assess the user's attention level and divide the user into tiers; The training task matching module matches the user with a training task of a corresponding difficulty level from the training question bank according to the attention level evaluation result and the user level; The training effect evaluation module identifies the user's performance data on the training task, performs automatic analysis on it, and generates an evaluation report; The attention judgment module determines whether the user's attention has changed through the evaluation report. If not, the training at the current difficulty level will continue. If there is a change, the module will return to re-evaluate and match the training task.

8. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the function of the system according to any one of claims 1 to 6 is realized.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the functions of the system according to any one of claims 1 to 6 are realized.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program realizes the functions of the system according to any one of claims 1 to 6.

Images