College student labor practice education system based on situation embedding

Abstract

The invention relates to the field of education technology and intelligent systems, and discloses a college student labor practice education system based on situation embedding, comprising a situation adaptation optimization engine for evaluating student ability states and dynamically adjusting task difficulty parameters; the personalized task pushing module customizes a differentiated task list and a learning path; the real-time evaluation system generates behavior performance scores and immediate feedback; the adaptive situation generation module optimizes the follow-up task situation complexity; and the system and platform interface synchronously presents task progress and learning analysis reports. According to the method, the problems of capability mismatching and feedback lagging caused by traditional static layering are solved through real-time expression score driving task difficulty dynamic adaptation and in combination with a behavior characteristic feedback immediate adjustment strategy; synchronously constructing a situation complexity regulation and control mechanism; and a teacher adjustable interface is integrated to realize a teaching closed loop, so that the teacher deeply participates in path adjustment and optimization. Four-dimensional cooperation breaks through the limitation of a fixed scene, and the degree of participation and the migration ability are enhanced while the learning efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of educational technology and intelligent system technology, and specifically to a labor practice education system for college students based on context embedding. Background Art

[0002] In the current higher education system, the value of practical labor education, as a crucial component of the moral education system, is increasingly recognized. Especially in light of educational evaluation reforms, enhancing the practicality, contextuality, and educational value of labor education has become a core issue facing universities. Traditional labor courses often focus on task assignments and achievement presentations, lacking effective process guidance and making it difficult to objectively quantify teaching effectiveness. While some universities have experimented with introducing digital platforms to assist with management, overall, they still prioritize post-process documentation over ongoing intervention.

[0003] Some educational information technology platforms have already introduced task-driven and behavioral data collection mechanisms to support the management of labor education activities. These systems typically assign tasks through static task lists, supplemented by scoring mechanisms to document performance. Some platforms also integrate simplified learning analytics modules. Teachers can review student submissions and conduct manual evaluations. The system can also provide simple categorized push notifications based on preset tags. However, the matching of tasks to student abilities is limited, the path planning logic is crude, and it lacks sufficient adaptive features, making it difficult to effectively guide students through their staged growth.

[0004] The key shortcoming of the existing labor practice education system is its failure to form a closed-loop feedback loop. Task recommendations are mainly based on static rules and lack real-time response to students' current status, resulting in frequent problems of "too difficult" or "too easy" tasks being pushed. Path recommendations are too template-based, ignoring individual differences among students, are highly repetitive, and lack growth logic. Although real-time data is collected, it is not used for dynamic intervention, and the system behavior lacks "strategy memory", resulting in a break in the continuous learning experience. It is also difficult for teachers to grasp the changes in students' abilities in real time and form effective intervention strategies. The disconnect between teaching and assessment still exists. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a labor practice education system for college students based on context embedding, which solves the problems in the existing labor practice education system such as the disconnection between task push and student capabilities, templated path recommendation, delayed process feedback, and lack of real-time teacher intervention.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a labor practice education system for college students based on context embedding, comprising:

[0007] A context-adaptive optimization engine that assesses students' current ability status based on their real-time behavioral data during labor tasks and generates task difficulty adjustment parameters;

[0008] A personalized task push and path design module is used to receive the task difficulty adjustment parameters and generate matching personalized tasks and learning paths based on the student's historical behavior data;

[0009] Real-time evaluation and dynamic feedback system, used to monitor students' real-time behavior data and provide performance scores and feedback information;

[0010] An adaptive scenario generation and adjustment module, configured to receive the performance score and feedback information and dynamically adjust the scenario complexity of subsequent tasks;

[0011] The system and platform interface is used to connect teachers and students, display task progress, learning feedback and personalized learning reports, and receive teaching intervention information input by teachers for reference by the context adaptation optimization engine.

[0012] Preferably, the context-adaptive optimization engine analyzes students' behavioral data through deep learning, calculates students' performance scores in tasks, and adjusts task difficulty based on the scores to ensure that the challenge of the tasks is appropriate to students' abilities.

[0013] Preferably, the personalized task push and path design module includes:

[0014] Task recommendation unit, used to push personalized tasks based on students' historical task performance and real-time feedback;

[0015] The path dynamic adjustment unit is used to dynamically adjust the learning path based on the student's real-time performance score to ensure that the task difficulty increases gradually;

[0016] The personalized task generation unit calculates and recommends tasks using the following formula:

[0017]

[0018] Among them, w i is the weight coefficient of task i, A t,i Score the student's performance on task t, P t+1 Rating for the recommended tasks.

[0019] Preferably, the path dynamic adjustment unit adjusts the student's learning path according to the following formula:

[0020] S t+1 =S t +λ·(P t+1 -S t );

[0021] Among them, S t is the student’s current learning status, P t+1is the score of the recommendation task, and λ is the path adjustment coefficient.

[0022] Preferably, the task recommendation unit adapts to the task difficulty E provided by the context optimization engine. t+1 Personalized task push is carried out to ensure that tasks always adapt to students' current learning level, avoiding underestimation of students' abilities or excessive challenges.

[0023] Preferably, the real-time evaluation and dynamic feedback system includes:

[0024] Task monitoring unit, used to monitor students' task execution process in real time and collect relevant behavioral data;

[0025] The evaluation calculation unit is used to calculate the student's task performance score B using the following formula t :

[0026]

[0027] Among them, T t is the actual completion time of the task, T max =600: Maximum allowed time for the task; D t : Difficulty of current mission target; D max =1.0: Maximum task difficulty defined by the system; Accuracy t Efficiency is the accuracy of the task. t is the efficiency of the task; α1, α2, α3, α4 are the weights of the scoring items.

[0028] The instant feedback generation unit generates instant feedback suggestions based on student performance to help students improve their operation strategies.

[0029] Preferably, the instant feedback generating unit generates improvement suggestions based on the student's task performance and calculates the feedback score using the following formula:

[0030] F t+1 =β·(B t -B target );

[0031] Among them, F t+1 Score the generated feedback, B t Score students' real-time performance, B target is the preset ideal task completion score, and β is the feedback adjustment coefficient.

[0032] Preferably, the adaptive scenario generation and adjustment module includes:

[0033] A situation complexity calculation unit is used to calculate the situation complexity of the current task based on the student's performance in the task;

[0034] The context adjustment unit is used to dynamically adjust the complexity of the task context according to the failure rate of the student task and the task execution status to ensure that the task is adapted to the student's ability level.

[0035] Preferably, the system and platform interface includes:

[0036] Task management unit, used by teachers to set and adjust task difficulty and task type;

[0037] Progress tracking unit, used to track students' learning progress in real time and display task completion status;

[0038] The learning report generation unit is used to generate personalized learning reports based on students' task performance and provide feedback to teachers to optimize the teaching process.

[0039] The present invention also provides a labor practice education method for college students based on context embedding, comprising the following steps:

[0040] S1. Calculate students' task completion scores based on their real-time performance data and adjust the difficulty of the task to suit the students' current abilities.

[0041] S2: Based on students' historical task data, personalized needs, and real-time performance scores, we push appropriate tasks and design personalized learning paths.

[0042] S3, monitor students' task execution process in real time, generate immediate feedback by calculating performance scores, and adjust learning strategies based on the feedback;

[0043] S4. Dynamically adjust the task context complexity based on student performance to ensure that the task difficulty is appropriate for students’ learning progress;

[0044] S5. Teachers adjust the task difficulty and track students’ learning progress through the system interface, and generate personalized learning reports for reference.

[0045] The present invention provides a labor practice education system for college students based on context embedding. It has the following beneficial effects:

[0046] 1. This invention utilizes a dynamic task recommendation mechanism driven by real-time performance scores, achieving precise matching of task difficulty and personalized path generation. Unlike existing solutions that rely on static ability-based tiered recommendations, this approach prevents students from chronically overestimating or underestimating tasks, addressing the problem of low learning efficiency. In practical applications, it consistently matches students' ability curves, maintaining optimal cognitive challenge ranges.

[0047] 2. This invention builds a real-time feedback model driven by behavioral characteristics, dynamically generating instant scores and adjusting task strategies during task execution. Unlike traditional models that rely solely on outcome feedback, this system can intelligently intervene during the learning process. This not only alleviates the frustration caused by overly difficult tasks but also addresses the adaptive limitations of delayed feedback, making learning strategies more intelligent.

[0048] 3. The invention proposes a mechanism for dynamic control of contextual complexity, adjusting the level of task information interference and the density of contextual elements in real time based on students' current performance. This breaks through the limitations of fixed task scenarios and single learning paths, truly achieving "customized questions." This ensures that students are consistently challenged at a level that's neither too easy nor too overwhelming, effectively improving their engagement and transferability.

[0049] 4. This invention integrates task management, path generation, and teacher interface modules to provide an adjustable and visual task scheduling solution. Compared to traditional systems where teachers passively monitor progress, this system enhances teacher intervention and strategic flexibility. It allows teachers to truly participate in the optimization of personalized learning paths, moving beyond simply observing results to adjusting the process. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a system structure diagram of the present invention;

[0051] Figure 2 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0053] Please see the attached Figure 1 The embodiment of the present invention provides a labor practice education system for college students based on context embedding, including:

[0054] A context-adaptive optimization engine that assesses students' current ability status based on their real-time behavioral data during labor tasks and generates task difficulty adjustment parameters;

[0055] The contextual adaptation optimization engine is one of the core decision-making modules in this system. Its function is to dynamically evaluate students' current ability status based on their real-time behavioral data during the execution of labor tasks, and adjust the difficulty level of subsequent tasks accordingly to achieve an adaptive match between individual abilities and task challenges.

[0056] The engine mainly includes the following physical structure components:

[0057] Behavioral data collection unit, which is connected to the real-time evaluation and dynamic feedback system to obtain basic behavioral data of students during task execution;

[0058] an ability status assessment unit, which is connected in data communication with the behavior data collection unit and is used to score the student behavior data;

[0059] A task difficulty adjustment unit, which is logically connected to the capability status evaluation unit and outputs a task difficulty parameter;

[0060] The parameter transmission interface unit communicates with the task recommendation unit in the personalized task push and path design module to achieve optimized parameter output.

[0061] During the system initialization phase, the platform sets the following initial parameters:

[0062] T max : The maximum completion time of the task, in seconds. The system default setting is 600 seconds;

[0063] D max : The maximum difficulty value of the task, the unit is a dimensionless fraction, and the value range is [0, 1];

[0064] α1=0.25、α2=0.35、α3=0.25、α4=0.15:Score weights, satisfying ∑α i =1.

[0065] After each round of tasks is completed, the system calculates the student's ability score B based on the following behavioral data t :

[0066]

[0067] in:

[0068] T t : actual task completion time, in seconds;

[0069] Accuracy t : Task completion accuracy, defined as the number of correct items completed divided by the total number of tasks, with a value range of [0, 1];

[0070] Efficiency t : The efficiency of completing the task per unit time;

[0071] D t : The preset difficulty value of the current task, initialized and generated by the task recommendation unit;

[0072] B t: Current ability score, with a value range of [0, 1], is used to measure the student's comprehensive performance in this round of tasks.

[0073] The capability status assessment unit further introduces a historical performance smoothing function to score B in the last n stages. t Perform weighted averaging to suppress occasional error fluctuations and enhance scoring stability:

[0074]

[0075] in:

[0076] Weighted smoothed score; w i : Scoring weight within the window, satisfying ∑w i =1; preferably using exponential decay form Where ρ∈(0,1), the default value is ρ=0.7;

[0077] n: The length of the history window, preferably set to 3-5 rounds.

[0078] After the score is output, the task difficulty adjustment unit will reversely adjust the target difficulty E of the subsequent task according to the score. t+1 , the adjustment formula is as follows:

[0079]

[0080] in:

[0081] γ: Response sensitivity coefficient, which controls the adjustment of score fluctuations to task difficulty. The default value is 0.8;

[0082] δ: Minimum baseline difficulty of the task, which prevents the task difficulty from falling to the point where it cannot be learned. The default setting is 0.1;

[0083] E t+1 : The target difficulty value of the next round of task recommendation is output to the task recommendation unit to drive task screening;

[0084] When students' ability scores are low (e.g. ), then the obtained E t+1 Reduce, effectively reduce the difficulty of the next round of tasks; on the contrary, when the score is high (such as ), the task difficulty will be increased accordingly to provide a challenge.

[0085] The parameter transmission interface unit transmits the task difficulty parameter E to the target object through a standard data interface (such as WebSocket or HTTP RESTful interface). t+1 Output, written into the task parameter field of the personalized task push module.

[0086] A personalized task push and path design module is used to receive the task difficulty adjustment parameters and generate matching personalized tasks and learning paths based on the student's historical behavior data;

[0087] The personalized task push and path design module is set up in the server-side decision-making layer, physically interconnected with the context adaptation optimization engine and real-time feedback system, logically relying on the scoring results and task difficulty parameters as input conditions, and outputting the individual task list and stage learning path status.

[0088] This module includes a task recommendation unit, a path dynamic adjustment unit, and a personalized task generation unit. The three are connected through an asynchronous data channel and are responsible for scoring judgment, path evolution control, and final task implementation generation, respectively.

[0089] The task recommendation unit is used to The task item that best matches the current student status is retrieved and scored based on multiple behavioral characteristics. The recommended score uses the following weighted model:

[0090]

[0091] in:

[0092] The student completes task T in round t+1. j preference ratings;

[0093] The student's behavior dimension i in the current round of task recommendation is related to task T j The adaptation value of

[0094] w i : The weight of the i-th behavioral dimension, satisfying

[0095] N: the total number of behavioral feature dimensions, i.e., i∈{efficiency, accuracy, time matching, path fitness}, in this system N=4;

[0096] The adaptation values of each dimension are defined as follows:

[0097] Efficiency dimension: Among them E j For task T j The target efficiency per unit time, Eff t is the historical average efficiency of students;

[0098] Accuracy dimension: Among them A j For task T j Target accuracy;

[0099] Time matching: Where T j The time to complete the task target;

[0100] Path fitness: Among them S t is the student’s current path status, D j The difficulty of the task.

[0101] The weight vector w = [w1, w2, w3, w4] can be set by the platform administrator or dynamically updated using the gradient descent method based on historical task completion results.

[0102] After the recommendation score is calculated, the task recommendation unit will filter the tasks. The filtering conditions are based on the target difficulty parameter E. t+1 The original difficulty of the task is D j The deviation is controlled within the tolerance range ∈, and the expression is:

[0103] |D j -E t+1 |≤∈;

[0104] in:

[0105] E t+1 : The target difficulty output by the context-adaptive optimization engine;

[0106] ∈: The system default setting is 0.1, ensuring that task screening is both stable and responsive;

[0107] D j :Task T j The difficulty value range is [0,1].

[0108] The personalized task generation unit will generate all tasks that meet the conditions The top k recommended tasks are pushed to the client. The task information structure includes task identification, description, difficulty level, target accuracy and recommendation reason.

[0109] The path dynamic adjustment unit is used to maintain and update the student's progress status in the entire task sequence. The path status is defined using a sliding weighted average method, and the initial state is:

[0110]

[0111] The system updates the path status S after each task push is completed t :

[0112] S t+1 =S t +λ·(P t+1 -S t );

[0113] in:

[0114] S t+1 : New path status, reflecting the student's ability adaptation trajectory;

[0115] P t+1 : The recommended score value of the current task (which can be approximated as the actual score B after the task is completed) t+1 );

[0116] λ: Path adjustment coefficient, the default value is 0.2, which indicates the learning path response speed.

[0117] The path status is fed back to the context adaptation optimization engine and recommendation unit as a weighted input for path fitness calculation and score update in subsequent task screening.

[0118] The task recommendation and path control modules are both connected to the back-end database and front-end system asynchronously, and data transmission is completed using HTTPS API or MQ protocol.

[0119] During actual deployment, the recommendation scoring, path status, and task screening processes are all completed in the local inference engine, with a response time of no more than 1 second. The lightweight model can be deployed by calling Python or C++.

[0120] The complete process is as follows: First, obtain the optimization engine score Then derive the target task difficulty E t+1 , and then the completed tasks are recommended, and the path status is dynamically updated, and finally fed back to the optimization engine to form a closed loop.

[0121] Through the design of this module, the system can provide high-frequency iterative task content push based on individual differences of students, and guide them to improve their capabilities along a dynamically evolving path, effectively avoiding learning stagnation or waste of resources.

[0122] Real-time evaluation and dynamic feedback system, used to monitor students' real-time behavior data and provide performance scores and feedback information;

[0123] The real-time evaluation and dynamic feedback system is set up in the middle layer of the platform. Its function is to collect and process the whole process of students' behaviors when performing labor tasks, complete ability scoring through mathematical models, and generate structured feedback, which is output to the personalized task push module and situational adaptation optimization engine.

[0124] The system includes a task monitoring unit, an evaluation calculation unit, and a feedback generation unit. These three units are connected through a unified message queue and establish data communication channels with the user client, database system, and optimization engine.

[0125] The task monitoring unit consists of the following components:

[0126] The behavior collector collects the following raw data through the client front-end component:

[0127] Task start time t start and end time t end , calculate T t =t end -t start , in seconds;

[0128] User click and slide event stream E = {e1, e2, ...}, used to analyze the operation path;

[0129] Error count e num and the total number of task items n total ;Number of completed items n done .

[0130] The preprocessing module structures the above data into:

[0131] T t : Task completion time;

[0132] Completion accuracy;

[0133] Completion rate per unit time.

[0134] System setting: Maximum task completion time T max 600 seconds; maximum difficulty D max =1.0.

[0135] After the collection is completed, the evaluation calculation unit will use a unified scoring model:

[0136]

[0137] The meanings of the parameters remain the same:

[0138] T t : actual completion time;

[0139] T max =600: maximum allowed time for the task;

[0140] Accuracy t : student completion accuracy, value range [0,1];

[0141] Efficiency t : The number of items completed per second, in items / second;

[0142] D t : The difficulty of the current task target;

[0143] D max=1.0: Maximum task difficulty defined by the system;

[0144] B t : Scoring result, range [0,1];

[0145] α1=0.25、α2=0.35、α3=0.25、α4=0.15:Rating item weights, satisfying ∑α i =1.

[0146] Scoring parameters for different task types can be differentiated through preset templates, such as increasing the weight of α3 for operational tasks and increasing the weight of α2 for cognitive tasks.

[0147] After the evaluation calculation is completed, the system will use the historical weighted average model to generate a smoothed score:

[0148]

[0149] in:

[0150] Smoothed historical average rating;

[0151] B t-i : the score of the ti-th round task;

[0152] n=3: scoring window length;

[0153] Exponential decay weight, ρ = 0.7, satisfies ∑w i =1.

[0154] The instant feedback generation unit generates a feedback response value according to the following deviation function:

[0155] F t+1 =β·(B t -B target );

[0156] in:

[0157] F t+1 : Feedback score, the range can be greater than [-1, 1]; β = 1.5: Feedback response sensitivity coefficient; B target =0.75: The ideal scoring threshold defined by the system.

[0158] According to F t+1 The value range is used to classify the feedback content:

[0159] If F t+1 <-0.1, generates negative improvement suggestions, such as "It is recommended to adjust the operation sequence to improve the completion accuracy";

[0160] If |F t+1|≤0.1, generates neutral suggestions, such as "The completion of this round of tasks is close to the standard, keep working hard";

[0161] If F t+1 >0.1, generates positive prompts, such as "The task is completed efficiently, and it is encouraged to maintain the current pace."

[0162] The feedback suggestions are sent to the user interface in text form, written into the database and cache layer as structured data, and simultaneously pushed to the task recommendation module as a basis for the next round of task candidates.

[0163] The data transmission structure defines the field format and data type in the interface specification to ensure compatibility and real-time performance between system modules.

[0164] Rating result B t With smoothing score It will be written into the context adaptation optimization engine in real time and used to predict the difficulty of subsequent tasks, which is determined by the following model:

[0165]

[0166] Where γ = 0.8 and δ = 0.1. The model has been fully defined in the optimization engine module.

[0167] The system is deployed using a service-oriented architecture, with each module running independently in a Docker container and utilizing a RESTful API or WebSocket bidirectional push mechanism. Scoring latency is controlled within 0.5 seconds, and feedback output is no more than 1 second, meeting real-time system response standards.

[0168] In summary, this system has built a complete technical chain from behavior collection, indicator extraction, ability scoring, history smoothing to feedback generation, data writing and task linkage, supporting students' real-time evaluation and dynamic optimization during the labor task process, and providing an efficient feedback basis for task push and path design.

[0169] An adaptive scenario generation and adjustment module, configured to receive the performance score and feedback information and dynamically adjust the scenario complexity of subsequent tasks;

[0170] The adaptive context generation and adjustment module is used to model and evaluate the adaptation status reflected by the student's current task performance, and based on the evaluation results, adjust the intensity of elements in subsequent tasks in terms of context, prompts, visual complexity, etc., thereby constructing a personalized task context environment.

[0171] This module is set in the system service middle layer and includes a situation complexity calculation unit, a situation control function construction unit and a task generation feedback interface unit. The three units are logically connected in sequence and physically establish a two-way data channel with the following modules: real-time evaluation and dynamic feedback system, personalized task push module, and situation adaptation optimization engine.

[0172] The situation complexity calculation unit receives the task performance score result B t and task execution status data, including:

[0173] T t : The completion time of the current round of tasks, in seconds;

[0174] Accuracy t : Completion accuracy, defined as

[0175] n done : The number of completed tasks, in items;

[0176] e num : The number of error items, the unit is item;

[0177] D t : Current task difficulty (provided by the task push module);

[0178] T max : The system sets the maximum allowed completion time, which is fixed at 600 seconds.

[0179] And calculate the failure rate F of the task execution results based on the following model t :

[0180]

[0181] The normalized time consumption index is:

[0182]

[0183] Task failure rate and task completion time T t Together they constitute the situational adaptability index C of task performance t , as shown below:

[0184] C t =ω1·R t +ω2·F t ;

[0185] in:

[0186] C t : Situational adaptation index of task execution;

[0187] ω1, ω2: Task performance influencing factor weights. The default values are ω1 = 0.5, ω2 = 0.5.

[0188] The C t The index reflects the degree of students' adaptability to the current task situation. The higher the value, the worse the student's performance, and the complexity of the task situation should be adjusted to reduce it.

[0189] To adapt to different task types, the system adjusts the weights based on the task labels. For tasks with high operational requirements, ω2 is increased to 0.7; for tasks with high time pressure, ω1 is increased to 0.6.

[0190] The above C t The context control function is sent to the construction unit, and the expected context adaptation value C defined by the system is used. target =0.3 for dynamic adjustment:

[0191] ΔQ t =μ·(C t -C target );

[0192] in:

[0193] ΔQ t : The complexity adjustment of the current task situation;

[0194] μ: Control sensitivity coefficient, the system is set to 0.15;

[0195] C target =0.3: Indicates the medium level of adaptation that the system expects students to maintain.

[0196] The situation adjustment unit is used to adjust the current task situation parameters according to the C t The value adjusts the situational complexity level of the next round of tasks. t+1 The expression is as follows:

[0197] Q t+1 =clip(Q t +ΔQ t ,Q min ,Q max );

[0198] in:

[0199] Q t : The complexity of the current task situation;

[0200] Q t+1 : Adjusted situational complexity;

[0201] Q min =0.1,Q max=0.9: minimum and maximum situational complexity allowed by the system;

[0202] clip(·): is a function that limits the range of output values.

[0203] Situational complexity Q t+1 It will be broken down into multiple task generation parameters to control the specific strength of different situational elements in subsequent task construction:

[0204] Visual complexity parameter Control the number of page prompt colors and auxiliary image density;

[0205] Prompt frequency parameter H t+1 =1-Q t+1 : Control whether to enable voice / step prompts;

[0206] Text precision parameter L t+1 =1-|0.5-Q t+1 |: controls the semantic ambiguity of task description;

[0207] Task context length Controls the event step size of the task context settings.

[0208] After receiving the above parameters, the task generation module will automatically match the preset context script template to construct a personalized task context that is suitable for the student's current performance.

[0209] In addition, the context complexity will also be used as a variable to generate the target task difficulty of the optimization engine, and the following model will be updated:

[0210]

[0211] The definitions of the variables are the same as before:

[0212] Historically weighted ratings;

[0213] γ=0.8、δ=0.1:rating influence coefficient;

[0214] ξ=0.1: indirect amplification factor of situation complexity on task difficulty;

[0215] Q t+1 : The complexity result of the currently generated situation.

[0216] By putting Q t+1 By controlling the task context load intensity and participating in the task difficulty update, a closed-loop optimization process based on task structure-feedback results-learning path can be constructed.

[0217] The execution process of this module is as follows: first, the failure rate is constructed based on the student task completion behavior data; then, the adaptation index is generated based on the time consumption; finally, the complexity of the task situation is adjusted according to the adaptation index, and the results are fed back to the task push and optimization engine.

[0218] Through the implementation of this module, an adjustment mechanism for the "contextual intensity" dimension can be provided in addition to task difficulty, supporting intelligent adjustment of task text, images, prompts and interactive details, ensuring that students remain in the "achievable but slightly challenging" cognitive range, and building a dynamically balanced labor education learning environment.

[0219] The system and platform interface is used to connect teachers and students, display task progress, learning feedback and personalized learning reports, and receive teaching intervention information input by teachers for reference by the context adaptation optimization engine.

[0220] The system-platform interface module is a core component of the entire system, responsible for data exchange, task allocation, task recommendation, progress tracking, and feedback output between submodules. This interface provides a data exchange channel through a standardized API, allowing real-time communication of student task performance, generating personalized learning paths, and providing real-time feedback to the task recommendation module.

[0221] The system and platform interface module includes the following units:

[0222] Task management unit: responsible for receiving task recommendations and task difficulty information from the personalized task push module and the context adaptation optimization engine, updating the task progress after receiving the student's task completion status, and generating the next round of task recommendations.

[0223] Progress tracking unit: responsible for real-time tracking of students' task completion progress in each task round and updating the global learning progress.

[0224] Learning report generation unit: Generates personalized learning reports based on task execution for teachers and students to use. The report includes the task score, status, completion time, learning progress, etc.

[0225] All these units communicate through a unified data interface standard, use JSON format for data transmission, and exchange real-time data through HTTP or WebSocket.

[0226] The task management unit is responsible for obtaining task difficulty and task candidate information from the optimization engine and task push module, and selecting appropriate tasks for push based on this information. Task screening uses the following calculation model:

[0227] Task filtering formula:

[0228]

[0229] in:

[0230] Task T j Recommendation score in the current round t+1;

[0231] The student's adaptation value for the i-th behavioral dimension in the history task;

[0232] w i : The weight coefficient of the behavioral dimension, satisfying

[0233] N: The total number of behavioral dimensions.

[0234] This formula prioritizes task recommendations by evaluating students' historical task data, sorting them from high to low based on their scores, and selecting the tasks that best suit the student's current ability level. Once the task recommendation list is generated, the task management unit delivers it to the student, waiting for them to complete it.

[0235] The progress tracking unit monitors the completion status of students' tasks in real time and calculates their learning progress. It calculates the overall progress of task completion based on the following formula:

[0236]

[0237] in:

[0238] S t : The student's task completion progress in round t;

[0239] status i : The completion status of task i (0 means unfinished, 1 means completed);

[0240] n: The total number of current tasks.

[0241] The progress tracking unit is responsible for feeding back the task status to the task management unit in real time so that the difficulty of the task and the recommended strategy can be adjusted in the next round. At the same time, it transmits the student's overall progress data to the learning report generation unit.

[0242] The learning report generation unit generates a detailed personalized learning report based on the student's task execution status, scoring results and learning progress.

[0243] The learning report generation unit stores the report in the database and makes it available to teachers and students for review through the platform interface.

[0244] All data is transmitted via a standardized RESTful API interface in JSON format, which complies with Web service standards. Each interface request and response includes information such as the corresponding task data, learning progress, and scoring results.

[0245] The system and platform interface uses a JSON-based data format, and all data interactions use HTTP / HTTPS protocols for requests and responses. The format of interface requests and responses strictly follows the following standards:

[0246] Task recommendation interface: used to pass task recommendation results to students. The interface returns information such as the difficulty and task description of the recommended task.

[0247] Task completion status interface: used to receive the status of tasks completed by students. After the task is completed, the task status (success, failure) is sent to the server.

[0248] Learning progress interface: real-time tracking and uploading of students' learning progress, helping the system adjust task recommendation strategies based on students' learning situation.

[0249] Learning report generation interface: Generate and push personalized learning reports for teachers and students to review.

[0250] Each interface provides a response timeout mechanism to ensure high availability of the system.

[0251] The system and platform interface utilizes a microservices architecture, breaking down functions such as task management, progress tracking, and learning reports into independent services, each deployed in a separate container. These services interact with the backend database via message queues (such as Kafka and RabbitMQ), ensuring efficient communication and data consistency between modules. Real-time performance is ensured through an asynchronous data transmission mechanism. At the end of each task round, the interface completes the entire process of task recommendation, status update, and report generation within 1 second.

[0252] The context-embedded labor practice education method for college students described below and the context-embedded labor practice education system for college students described above can be used in conjunction with each other.

[0253] See also Figure 2 The present invention also provides a labor practice education method for college students based on context embedding, comprising the following steps:

[0254] S1. The system collects real-time student performance data during tasks, including completion time, error rate, and click behavior, to calculate a task completion score. The score is compared with the preset task objectives to dynamically assess the student's current ability level and fine-tune the difficulty of subsequent tasks to achieve a proper match between difficulty and ability.

[0255] S2: The system performs a multi-dimensional analysis of students' historical task data and personalized parameters (such as learning preferences and weaknesses). Combined with the current scoring results, it generates a set of candidate tasks. It then prioritizes these tasks based on their matching scores and recommends the most suitable tasks, building a personalized learning path that aligns with cognitive load and ability progression.

[0256] During task execution, the system monitors and analyzes behavioral characteristics in real time, such as dwell time, operation sequence, and feedback response. By comparing performance with the model's expected behavior, it generates an immediate performance score and uses this score to adjust the current strategy, such as interrupting, simplifying, or replacing the task, achieving dynamic strategic response.

[0257] S4. Task context complexity is controlled through multi-dimensional variables, including the number of task context elements, the degree of information interference, and response constraints. The system dynamically adjusts these parameters based on the student's current performance to ensure that the task is challenging but not excessively burdensome, ensuring the continuity and adaptability of the learning process.

[0258] The S5. Teacher-side interface provides real-time task distribution and student progress query capabilities. Teachers can manually intervene in task design by adjusting task parameters. The system also automatically generates learning reports based on students' periodic task execution data. The reports include progress overviews, score trends, and key feedback to support precise teaching decisions.

[0259] The device of this embodiment can be used to execute the above method embodiment, and its principles and technical effects are similar, so they will not be repeated here.

[0260] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. The labor practice education system for college students based on context embedding is characterized by: include: A context-adaptive optimization engine that assesses students' current ability status based on their real-time behavioral data during labor tasks and generates task difficulty adjustment parameters; A personalized task push and path design module is used to receive the task difficulty adjustment parameters and generate matching personalized tasks and learning paths based on the student's historical behavior data; Real-time evaluation and dynamic feedback system, used to monitor students' real-time behavior data and provide performance scores and feedback information; An adaptive scenario generation and adjustment module, configured to receive the performance score and feedback information and dynamically adjust the scenario complexity of subsequent tasks; The system and platform interface is used to connect teachers and students, display task progress, learning feedback and personalized learning reports, and receive teaching intervention information input by teachers for reference by the context adaptation optimization engine.

2. The context-embedded labor practice education system for college students according to claim 1 is characterized in that: The context-adaptive optimization engine analyzes students' behavioral data through deep learning, calculates students' performance scores in tasks, and adjusts task difficulty based on the scores to ensure that the challenge of the tasks is appropriate to the students' abilities.

3. The context-embedded labor practice education system for college students according to claim 1 is characterized in that: The personalized task push and path design module includes: Task recommendation unit, used to push personalized tasks based on students' historical task performance and real-time feedback; The path dynamic adjustment unit is used to dynamically adjust the learning path based on the student's real-time performance score to ensure that the task difficulty increases gradually; The personalized task generation unit calculates and recommends tasks using the following formula: Among them, w i is the weight coefficient of task i, A t,i Score the student's performance on task t, P t+1 Rating for the recommended tasks.

4. The context-embedded labor practice education system for college students according to claim 3 is characterized in that: The path dynamic adjustment unit adjusts the student's learning path according to the following formula: S t+1 =S t +λ·(P t+1 -S t ); Among them, S t is the student’s current learning status, P t+1 is the score of the recommendation task, and λ is the path adjustment coefficient.

5. The context-embedded labor practice education system for college students according to claim 3 is characterized in that: The task recommendation unit adapts the task difficulty E provided by the context optimization engine to the task t+1 Personalized task push is carried out to ensure that tasks always adapt to students' current learning level, avoiding over-challenging or underestimating students' abilities.

6. The context-embedded labor practice education system for college students according to claim 1 is characterized in that: The real-time evaluation and dynamic feedback system includes: Task monitoring unit, used to monitor students' task execution process in real time and collect relevant behavioral data; The evaluation calculation unit is used to calculate the student's task performance score B using the following formula t : Among them, T t is the actual completion time of the task, T max =600: Maximum allowed time for the task; D t : Difficulty of current mission target; D max =1.0: Maximum task difficulty defined by the system; Accuracy t Efficiency is the accuracy of the task. t is the efficiency of the task; α1, α2, α3, α4 are the weights of the scoring items. The instant feedback generation unit generates instant feedback suggestions based on student performance to help students improve their operation strategies.

7. The context-embedded labor practice education system for college students according to claim 6 is characterized in that: The instant feedback generation unit generates improvement suggestions based on the student's task performance and calculates the feedback score using the following formula: F t+1 =β·(B t -B target ); Among them, F t+1 Score the generated feedback, B t Score students' real-time performance, B target is the preset ideal task completion score, and β is the feedback adjustment coefficient.

8. The context-embedded labor practice education system for college students according to claim 1 is characterized in that: The adaptive scenario generation and adjustment module includes: A situation complexity calculation unit is used to calculate the situation complexity of the current task based on the student's performance in the task; The context adjustment unit is used to dynamically adjust the complexity of the task context according to the failure rate of the student task and the task execution status to ensure that the task is adapted to the student's ability level.

9. The context-embedded labor practice education system for college students according to claim 1 is characterized in that: The system and platform interface include: Task management unit, used by teachers to set and adjust task difficulty and task type; Progress tracking unit, used to track students' learning progress in real time and display task completion status; The learning report generation unit is used to generate personalized learning reports based on students' task performance and provide feedback to teachers to optimize the teaching process.

10. A method for labor practice education for college students based on context embedding, applied to a labor practice education system for college students based on context embedding as described in any one of claims 1 to 9, characterized in that: The steps include: S1. Calculate students' task completion scores based on their real-time performance data and adjust the difficulty of the task to suit the students' current abilities. S2: Based on students' historical task data, personalized needs, and real-time performance scores, we push appropriate tasks and design personalized learning paths. S3, monitor students' task execution process in real time, generate immediate feedback by calculating performance scores, and adjust learning strategies based on the feedback; S4. Dynamically adjust the task context complexity based on student performance to ensure that the task difficulty is appropriate for students’ learning progress; S5. Teachers adjust the task difficulty and track students’ learning progress through the system interface, and generate personalized learning reports for reference.

Images