Education supervision object-oriented education system information interaction method

Abstract

The invention belongs to the field of data analysis and processing, and discloses an education supervision object-oriented education system information interaction method. Comprising the steps that a supervisor sets an information fusion pool, receives information of multiple information sources and distributes the information through a single transmission channel; setting a front pool, dynamically updating the intensity and the aging entropy value of the information, adjusting the attenuation speed of the information intensity according to user behaviors, and reminding a supervised person through color distinguishing display and sensory interaction; the supervisor performs structural decomposition on the information, defines a role pool and divides user groups, sets a game simulation pool to simulate an execution effect, generates a candidate execution path, updates the candidate execution path through a role switching rule, and generates an optimal execution path through information intensity dynamic update and path evaluation; and the supervisor quantifies the contribution degree of the user group based on the optimal execution path, generates optimization suggestions, collects feedback behaviors and distributes incentive values, thereby realizing education system information interaction oriented to the education supervision object.

Description

Technical Field

[0001] This invention relates to the field of data analysis and processing, and more specifically, to an information interaction method for educational systems oriented towards educational supervision targets. Background Technology

[0002] In the education oversight system, the distinction between regulators and those being regulated is typically based on administrative hierarchy and division of responsibilities. Regulators are those with decision-making, supervisory, and resource allocation authority, primarily responsible for policy formulation, issuing notices, supervising implementation, and resource management. Those being regulated are those responsible for implementing policies, completing tasks, and providing feedback, primarily responsible for receiving notices, performing tasks, submitting data, and reporting issues.

[0003] In the process of education supervision, the following problems exist in information exchange with those being supervised: During the information exchange process, the education system may frequently issue notices and reports, causing those being supervised to be overwhelmed by the massive amount of information and making it difficult to receive information effectively; In addition, the supervision involves multiple departments or personnel, which further complicates the information channels (such as WeChat groups, emails, dedicated platforms, etc.), thus exacerbating the information overwhelming phenomenon. Meanwhile, inconsistencies in information issued by multiple departments, unclear policy interpretations, and varying enforcement standards among different regulators can cause confusion among those being regulated. Furthermore, the dissemination of information through multiple channels makes it difficult for those being regulated to choose effective channels or entities to provide feedback or make requests, resulting in delayed responses and further deteriorating the information exchange process.

[0004] Therefore, information exchange within the education system targeting those subject to educational supervision needs to be designed and innovated to meet actual needs. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above objectives, the present invention provides the following technical solution: an information interaction method for an education system oriented towards educational supervision objects, comprising: S1: the supervisor sets up an information fusion pool, receives information from multiple information sources and distributes the information through a single transmission channel, and the supervised person uses multi-dimensional sensory interaction to distinguish, present and interact with the information; S2: The regulator sets up a pre-pool, dynamically updates the intensity and timeliness entropy of information, adjusts the decay rate of information intensity according to user behavior, and reminds the regulated by color differentiation and sensory interaction. S3: The regulator decomposes the information in a structured manner, defines the role pool and divides the user groups, sets up a game simulation pool to simulate the execution effect, generates candidate execution paths, updates the candidate execution paths through role switching rules, and generates the optimal execution path through dynamic updates of information intensity and path evaluation. S4: Regulators quantify user group contributions based on the optimal execution path, detect differences in contributions and generate optimization suggestions, predict deviation probabilities and trigger intervention mechanisms, collect feedback behaviors and allocate incentive values, record the allocation and redemption process of incentive values, allocate resources through automated execution mechanisms, dynamically adjust regulatory incentive values ​​and trigger intervention mechanisms, and optimize the feedback processing flow through path prediction models.

[0006] Preferably, S1 includes: Regulators set up an information fusion pool to receive information from multiple sources and add regulatory parameter clusters, which include the source identifier, priority identifier, and content summary of the information. The information is mapped to a uniform encoding format based on its source identifier and distinguished by the encoding value; then, the information is distributed through a single transmission channel. The regulatory parameter clusters for identifying information about the regulated entities are presented and interacted with using multi-dimensional sensory interaction. The regulator will transmit the information about adding regulatory parameter clusters to the front pool.

[0007] Preferably, S2 includes: For regulators, the initial intensity and initial time-limited entropy value of each information window are predefined according to the user role; A pre-pool is set up. After the information is identified and processed by the pre-pool, a uni-tripartite label (basic score | urgency, importance, timeliness) is added. The basic score is set according to urgency, importance, and timeliness. Identify the information type and fill it into the corresponding information window; Extract the average basic score of information within the window and dynamically update the intensity based on the information release frequency; extract the average urgency and importance of information within the window and dynamically update the updated time entropy value by combining the user's historical behavior and the initial time entropy value; customize the update cycle; The information intensity is calculated to decay naturally over time based on the time entropy value, and the decay rate is adjusted according to user behavior; the information intensity is updated according to the time frequency, and the information is displayed with different colors according to the threshold range it falls within.

[0008] Preferably, the method of updating the intensity of information according to time frequency and differentiating the display by color based on the threshold range includes: If the intensity of the current information is higher than the threshold, it is recorded as high intensity information, highlighted with a high-saturation color (such as orange or red) and flashed at 1-second intervals, accompanied by a 500-millisecond prompt sound and 3 consecutive short vibrations. If the intensity of the current information is below the threshold, it is recorded as low intensity information. Low intensity information reduces its transparency with the information intensity update time frequency and fades out after the time limit is exceeded. Before fading out, a single long vibration is used to remind the user.

[0009] Preferably, S3 includes; The regulator decomposes the information issued to the regulated party into multiple execution units based on the initial intensity and time entropy value of the information, including resource allocation, task division and execution time limit; Define a pool of roles and assign specific functionalities, objective functions, and resource constraints to each role. Users in education supervision are divided into multiple groups, each group corresponds to a role, and each role has specific decision-making rules; Set up a game simulation pool to simulate the execution effects of different roles and generate multiple candidate execution paths; Predefined role switching rules are used to evaluate the probability of role allocation based on the user's historical characteristic data. When the probability is higher than expected, role switching is triggered and an allocation plan is generated. Candidate execution paths are updated based on the allocation plan. Path convergence is achieved through dynamic updates of information intensity and path evaluation, and the optimal execution path is generated. Regulators collect feedback from those they regulate in order to optimize task allocation and generate the optimal execution path.

[0010] Preferably, S4 includes: Based on the generated optimal execution path, the contribution of each user group is quantified through a contribution evaluation mechanism; The system detects differences in contribution levels. When the differences exceed a preset threshold, it generates optimization suggestions and adjusts task allocation. The optimization suggestions are generated based on the dynamic role allocation of user groups. Predict the probability of deviation in task allocation after adjustment. When the probability of deviation exceeds a preset threshold, trigger the intervention mechanism and regenerate the optimal execution path. The intervention mechanism includes resource adjustment, strategy guidance and negotiation mechanism. Collect feedback behavior from the group and allocate entropy rewards. Feedback behavior includes reporting task allocation conflicts, confirming resource gaps, and proposing optimization solutions.

[0011] Preferably, the collection of feedback behavior from the group and the allocation of entropy rewards includes: Regulators set up a feedback incentive pool and allocate incentive values ​​based on the feedback behavior of the regulated entities. Feedback behavior includes reporting policy contradictions, confirming resource gaps, and proposing optimization suggestions. Resources are redeemed based on incentive values, which include automated task processing time, accelerated resource allocation channels, and regulatory exemption limits. Record the allocation and redemption process of incentive values, and realize resource allocation through an automated execution mechanism; The feedback status and incentive value balance are displayed through visualization and multimodal interaction.

[0012] Preferably, the feedback incentive pool is modeled as a virtual economic system, with the incentive value using entropy as the unit of measurement. 1 entropy value is exchanged for 1 minute of automated task processing time, 200 entropy values ​​are exchanged for a resource allocation acceleration channel, and 500 entropy values ​​are exchanged for a regulatory exemption quota.

[0013] Preferably, when the feedback behavior is not processed, the incentive value of the regulator is dynamically adjusted, and an intervention mechanism is triggered when the incentive value reaches a preset condition. The intervention mechanism includes notifying superiors and reassigning tasks. By using a distributed ledger to record feedback behavior, incentive value adjustments, and intervention processes, the records are formatted as key-value pairs; The path prediction model predicts the probability of deviation in feedback behavior. The input features of the path prediction model include the urgency of the feedback behavior, the scope of its impact, the historical processing time, and the pheromone intensity. The output is the probability of deviation in future feedback processing. When the prediction deviation probability exceeds a preset threshold, optimization suggestions are generated and the feedback processing flow is adjusted.

[0014] Preferably, the key-value pair record format for the feedback behavior is {User ID|Feedback Type|Processing Status|Timestamp}, the key-value pair record format for the incentive value adjustment is {Regulator ID|Deduction Entropy Value|Trigger Condition|Timestamp}, and the key-value pair record format for the intervention process is {Intervention Type|Intervention Object|Timestamp}.

[0015] The technical effects and advantages of the information interaction method for education systems oriented towards educational supervision objects of the present invention are as follows: By standardizing information from multiple sources through an information fusion pool, the problem of information channel clutter is eliminated, improving the identifiability and processability of information. A single transmission channel ensures efficient information delivery, preventing those being monitored from being overwhelmed by massive amounts of information and reducing the cognitive burden of information reception. Those being monitored can distinguish information sources and priorities through multi-dimensional sensory interaction; for example, high-priority information is presented with highlighted borders and vibration alerts, improving the efficiency of information reception and the accuracy of processing.

[0016] By employing an exponential decay model and adjusting user behavior, information self-purification is achieved, preventing low-value information from interfering with those being regulated and freeing up core teaching time. Based on predefined initial intensity and time-sensitivity entropy values ​​for user roles, the information needs of different roles are met. Information priority is intuitively presented through color differentiation and sensory interaction, reducing the information filtering costs for those being regulated.

[0017] By using structured decomposition, the specific content of policy implementation was clarified, reducing ambiguity in interpretation. Role pools and role switching rules improved the adaptability and efficiency of task allocation. A game theory simulation pool modeled the execution effects of different roles, resolving the issue of inconsistent execution standards among different regulators and ensuring fairness and efficiency in task allocation. The Shapley value quantified the contribution of each user group, objectively evaluating the implementation effect and addressing the problem of merely "implementing documents with documents" while neglecting actual implementation results.

[0018] By detecting differences in contribution, optimization suggestions are generated and task allocation is adjusted, improving the fairness and efficiency of implementation. Path prediction models are used to predict deviation probabilities, triggering intervention mechanisms in advance, reducing the risk of implementation deviations and ensuring the stability of policy implementation.

[0019] By allocating entropy rewards through a feedback incentive pool, regulated entities are encouraged to proactively report issues, thus improving both participation and quality. Exchanging entropy for resources reduces the workload for regulated entities and increases the practical value of feedback. The use of distributed ledgers and smart contracts to record and execute incentive allocation ensures transparency and efficiency in feedback processing, resolving the difficulty of selecting feedback channels.

[0020] By dynamically adjusting the incentive values ​​and intervention triggers for regulators, timely feedback is ensured, resolving issues of one-way communication and lack of closed-loop tracking. Through mechanisms such as dynamic role allocation, path prediction, and social feedback, the system's adaptability and scalability are enhanced, making it suitable for various educational supervision scenarios. Attached Figure Description

[0021] Figure 1 This is a schematic diagram illustrating the steps of an information interaction method for an education system oriented towards educational supervision objects, according to the present invention. Figure 2 This is a schematic diagram of the allocation process of the feedback excitation pool in this invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example

[0023] Please see Figure 1 and Figure 2 As shown in this embodiment, an information interaction method for education systems oriented towards educational supervision objects includes: In the design, regulators are typically the issuers of information and the formulators of policies. Those being regulated are typically the recipients of information and the implementers of tasks. It's important to note that in an education regulatory system, the roles of regulators and those being regulated can be dynamic, depending on the level and the task at hand. In the relationship between the education bureau and schools, the education bureau is the regulator, and the school (including the principal, teachers, and administrators) is the regulated. Within a school, the principal is the regulator, and teachers and administrators are the regulated. In certain tasks, teachers may simultaneously be both regulated (receiving tasks assigned by the principal) and regulators (supervising students in performing tasks).

[0024] In educational supervision scenarios, the initial intensity and time-sensitivity entropy of information should reflect its priority, urgency, and effective duration. The type of information window determines the role and priority of the information within it within the educational supervision system. High-priority information should have a higher initial intensity and a shorter time-sensitivity entropy value to quickly attract attention and ensure rapid transmission and processing; low-priority information should have a lower initial intensity and a longer time-sensitivity entropy value to avoid interfering with core tasks.

[0025] User behavior (such as read, ignored, no interaction) can serve as a basis for dynamically adjusting the initial intensity and time-dependent entropy value. For example, information types that are frequently ignored may require a lower initial intensity, while information types that are frequently read may require a longer time-dependent entropy value.

[0026] Regulators set up an information fusion pool to receive information from multiple sources (such as instant messaging tools, email systems, and regulatory platforms) and add regulatory parameter clusters, which include the source identifier, priority identifier, and content summary of the information. The information fusion pool acts as an information preprocessing center, standardizing information from multiple sources. For example, instant messaging messages might be transmitted as short texts, email messages as long texts, and regulatory platform messages might contain structured data. The information fusion pool uses natural language processing techniques (such as text classification and keyword extraction) or data parsing techniques to analyze the information and add multi-dimensional feature labels.

[0027] The specific implementation of the regulatory parameter cluster can include: source identifier (such as "IM" for instant messaging tools, "EM" for email systems, and "RP" for regulatory platforms), priority identifier (such as a value from 0 to 100, with 100 representing the highest priority), and content summary (such as the first 50 characters or keywords of the extracted information). For example, an urgent notification from an instant messaging tool might be labeled "IM|90|Urgent Meeting Notification".

[0028] The information is mapped to a uniform encoding format based on its source identifier and distinguished by the encoding value; then, the information is distributed through a single transmission channel. For example, source identifiers can be mapped to different "pheromone types" (e.g., instant messaging information is mapped to warning pheromone encoding, email system information to navigation pheromone encoding, and regulatory platform information to foraging pheromone encoding). The specific implementation of the encoding can be generated through hash functions or embedding vectors. For example, word embedding models (such as BERT) can be used to convert the information content into a high-dimensional vector, and then combined with source identifiers and priority identifiers to generate a unified encoding.

[0029] A single transmission channel can be implemented using message queue technologies (such as Kafka and RabbitMQ) or real-time data stream processing frameworks (such as Apache Flink) to ensure efficient distribution of information from multiple sources.

[0030] The regulatory parameter cluster identifies the information of the regulated entity and uses multi-dimensional sensory interaction to differentiate and interact with it.

[0031] When using multi-sensory interaction for differentiated presentation, information can be sorted according to priority indicators. For example, information with a priority indicator of ≥80 is placed at the top of the list. Information from different sources is distinguished by different visual styles and sensory cues. For example, information from instant messaging tools is presented with a green border and a short vibration (200 milliseconds), information from email systems is presented with a blue border and a long vibration (500 milliseconds), and information from regulatory platforms is presented with a purple border and a low-pitched alert tone (300 Hz).

[0032] Information sources can be filtered via voice commands, such as automatically filtering other sources and reading the filtered results after the user gives the command "only display information from the regulatory platform"; information sources can be switched via gestures, such as swiping left to switch to email system information and swiping right to switch to regulatory platform information; information priority can be reminded via tactile feedback, such as triggering three consecutive short vibrations on the smart bracelet for information with a priority indicator ≥90, and triggering a single long vibration for information with a priority indicator <60.

[0033] For regulators, the initial intensity and initial time-limited entropy value of each information window are predefined according to the user role; Users with different roles (such as teachers, principals, and administrators) have different information needs. For example, principals may be more concerned with policy interpretations, while teachers may be more concerned with emergency notices. Therefore, it is necessary to set personalized raw intensity and raw timeliness entropy values ​​for different roles.

[0034] Information windows are equivalent to predefined information categories based on the type of information. For example, information window types (such as emergency notices, routine reports, and policy interpretations). When a new piece of information is received, the type of information is determined, and then it is filled into the corresponding information window. All information in the information window uses the initial intensity and timeliness entropy value of the information corresponding to that window.

[0035] A pre-pool is set up. When each supervisor sends a message to a user (supervised user), the message is transmitted through the wireless network and first enters the pre-pool. After being identified and processed by the pre-pool, a one-to-three label (basic score | urgency, importance, timeliness) is added. The one-to-three label can be marked on the message. The basic score is set according to urgency, importance, and timeliness. The pre-processing pool can be configured with a machine learning model and a scoring mechanism. The machine learning model (e.g., neural networks, deep learning, or text recognition networks) is used to analyze and process the transmitted information, evaluating its urgency, importance, and timeliness. Urgency, importance, and timeliness can be represented numerically, where timeliness refers to the effective period of the information. For example, if a piece of information needs to be processed within two days, its timeliness is recorded as 2 days. Alternatively, the unit can be standardized and converted to minutes. The range of importance and timeliness values ​​can be standardized, typically set to 0 to 100, where 100 represents the highest level. The scoring mechanism can use a mathematical formula to define an evaluation function, such as: Base Score = w_1 × Urgency + w_2 × Importance + w_3 × Timeliness, where (w_1, w_2, w_3) are weighting coefficients, such as w_1 = 0.5, w_2 = 0.3, w_3 = 0.2.

[0036] Identify the information type and fill it into the corresponding information window; Extract the average basic score of information within the window and dynamically update the intensity based on the information's publication frequency; extract the average urgency and importance of information within the window and dynamically update the updated time entropy value by combining the user's historical behavior and initial time entropy value; customize the update cycle (e.g., one week or 10 days). Specifically, if the frequency f of a certain type of information is too high (such as the attendance registration form in a regular report being published multiple times a day), its initial intensity can be appropriately reduced to avoid information overload. Therefore, the initial intensity update mechanism can be defined as: Updated intensity = Average of base scores × 1 / (1 + log(f), where (f) is the frequency of this information's publication in past update cycles).

[0037] The time-sensitivity entropy value is the time required for the information intensity to halve, reflecting the effective duration of the information, and is measured in hours. An initial time-sensitivity entropy value is set based on the timeliness requirements of the information. Information with short timeliness (such as emergency notices) should have a shorter time-sensitivity entropy value, while information with long timeliness (such as policy interpretations) should have a longer time-sensitivity entropy value. Example of a time-sensitivity entropy value range: Urgent notice: 24 hours (requires prompt action).

[0038] Regular reports: 72 hours (medium time-sensitive).

[0039] Policy Interpretation: 240 hours (long-term reference).

[0040] Information of high importance but low urgency (such as policy interpretations) should have its timeliness entropy value extended to maintain long-term visibility. Information of high urgency but moderate importance (such as urgent notices) should have its timeliness entropy value shortened to highlight short-term priority. The timeliness entropy value should be dynamically adjusted based on user history. For example, if a certain type of information (such as routine reports) is frequently ignored by users, its timeliness entropy value can be shortened; if a certain type of information (such as policy interpretations) is frequently read by users, its timeliness entropy value can be extended.

[0041] For example, the updated timeliness entropy value = initial timeliness entropy value × (mean importance / mean urgency) × (1 + m × ((R − I) / (R + I))); R is the number of times the user has read the information within the update cycle, I is the number of times it has been ignored, and m is the adjustment coefficient, such as m = 0.2.

[0042] In educational oversight scenarios, the intensity of information (similar to pheromone concentration) is actively reduced based on the audience's cognitive load, and the time entropy value determines the decay rate. User behavior (such as ignoring or recognizing as read) can accelerate or slow down the decay rate, reflecting the actual value of the information and user needs.

[0043] The natural decay of information intensity over time is calculated based on the time-sensitivity entropy value, and the decay rate is adjusted according to user behavior. Specifically, the natural decay of information intensity over time follows a certain pattern. This pattern needs to be determined based on actual data analysis or expert evaluation. For example, if we use exponential decay as the law governing the natural decay of information intensity over time, we can design an exponential model: Q(t) = Q0 × e^(-t / t). −λt ==Q0×(1 / 2) t / T【1 / 2】 Where Q0 is the initial intensity, λ is the decay constant, T[1 / 2] is the half-life, and Q(t) is the information intensity at time t. The decay constant is calculated based on the initial intensity and half-life of the information, and then the time t is substituted into the information intensity Q(t) at time t.

[0044] Adjust the decay rate based on user behavior. For example, ignoring reduces the intensity of information by 50% and shortens the half-life to accelerate decay, while reading reduces the intensity of information by 75% and significantly shortens the half-life to accelerate decay. The intensity of information updates is determined by time frequency, and different colors are used to distinguish the information based on the threshold range. Different sensory interactions are also set to remind the monitored individuals.

[0045] The display status of information is adjusted according to its intensity. For example, high-intensity information (intensity ≥ 60) is highlighted with an orange or red border and flashes at 1-second intervals, accompanied by a 500-millisecond prompt tone and three consecutive short vibrations. Low-intensity information (intensity < 60) has its transparency reduced by 20% every 24 hours and fades out after 72 hours. Before fading out, it is reminded to the user with a single long vibration. High-intensity information can be read via voice commands, and the processing status of information can be marked by gestures. Swiping left marks it as read, swiping right marks it as pending, and swiping down marks it as archived or deleted. A list of low-intensity information to ignore is recommended based on the user's historical behavior, and users can confirm the batch ignoring with one click.

[0046] The implementation steps are illustrated below. Step A1: Input parameters: initial intensity, half-life, and current time t.

[0047] Step A2: Calculate the attenuation constant λ and the current intensity Q(t).

[0048] Step A3: Detect user behavior (ignore, read, no interaction) and the time t when it occurs.

[0049] Step A4: Calculate the adjusted information strength based on the user behavior type, ignoring behavior: Q 1 (t) = Q(t) × 0.5; Read lines Q 1 (t) = Q(t) × 0.25.

[0050] Step A5: Calculate the adjusted half-life based on the user behavior type, ignoring behavior T. 1 【1 / 2】 = T【1 / 2】×0.5; Read rows are T 1 【1 / 2】 = T【1 / 2】×0.1.

[0051] Step A6: Using the adjusted information strength and half-life, calculate the information strength at subsequent time (t').

[0052] Step A7: Update the display status of the information at the monitored entity based on the adjusted information intensity.

[0053] The information issued by the regulator to the regulated party is first decomposed into a structured form, generating multiple execution units, including resource allocation, task division and execution time limit; The specific implementation of structured decomposition can be achieved through text parsing techniques (such as dependency parsing) or expert knowledge bases, which decompose information text into multiple execution units. For example, a policy "The school needs to complete the equipment procurement next week" can be decomposed into: resource allocation (procurement budget), task assignment (procurement manager), and execution deadline (next Friday).

[0054] In educational oversight scenarios, the roles of user groups may dynamically change over time or for different tasks. For example, a teacher may assume an administrative role (such as a procurement manager) in some tasks, while a principal may assume a teaching role (such as curriculum evaluation) in others. Therefore, it is beneficial to allow user groups to dynamically switch roles based on task requirements during the negotiation process.

[0055] Define a pool of roles and assign specific functionalities, objective functions, and resource constraints to each role. Users in education supervision are divided into multiple groups, each group corresponds to a role, and each role has specific decision-making rules; A game simulation pool is set up to simulate the execution effects of different roles through a group game model. The group game model generates multiple candidate execution paths by defining the behavior model of the user group, constructing the game environment, and iteratively playing the game. The candidate execution paths are a set of policy execution trajectories generated by the group game.

[0056] The game simulation pool can be implemented using a multi-agent system, where each group acts as an agent, generating multiple candidate execution paths through a group game model. The implementation of the group game model includes the following steps: Define objective functions, resource constraints, and decision rules for each user group. For example, the functional behavior of the teacher role is "to perform teaching tasks," the objective function is "to maximize the fairness of task allocation," the resource constraint is "working time ≤ 40 hours / week," and the decision rule is "to prioritize tasks with the least workload." Define a hypothetical game environment, including the initial state of policy implementation (e.g., total resources, task list) and constraints (e.g., budget cap, time limit). Game rules include resource competition rules (e.g., first-come, first-served or proportional allocation) and task allocation rules (e.g., negotiation or forced assignment).

[0057] The decision-making process of user groups is simulated through multiple rounds of iteration. In each round, each user group adjusts its strategy based on its current state (the strategies of other groups and resource allocation) and its own objective function. For example, if the teacher group finds that tasks are unevenly distributed, it may adjust its strategy to "reject high-workload tasks"; if the principal group finds that resource utilization is low, it may adjust its strategy to "reallocate the budget". The game theory algorithm can use Nash equilibrium or reinforcement learning to calculate the strategy adjustment. For example, when using reinforcement learning, each user group acts as an agent, learning the optimal strategy through trial and error, and the reward function is the degree of achievement of the objective function (such as task fairness and resource efficiency).

[0058] Predefined role switching rules are used to evaluate the probability of role allocation based on the user's historical characteristic data. When the probability is higher than expected, role switching is triggered and an allocation plan is generated. Candidate execution paths are updated based on the allocation plan. Historical characteristic data, such as historical task completion rate, resource utilization rate, feedback behavior, and role preference, can be used. For example, if a teacher demonstrates high efficiency in procurement tasks (task completion rate ≥ 90%), the probability of them taking on an administrative role is predicted to be 0.85. When this probability is higher than a preset threshold (such as 0.8), a role switch is triggered.

[0059] After the role switch, a dynamic role allocation plan is generated, namely, "Teacher A switches to the administrative role and is responsible for procurement tasks; Principal B switches to the teacher role and is responsible for curriculum evaluation."

[0060] The candidate execution path is updated based on the dynamic role allocation scheme, such as adding a new path: "Administrative role teacher A centrally allocates procurement tasks, and teacher role principal B evenly allocates course evaluation tasks."

[0061] In group game models, the strategy space for role switching is expanded. For example, a group of teachers can choose not only to "distribute tasks evenly" but also to "switch to administrative roles and centrally distribute tasks." The role switching strategy is optimized through an information intensity update mechanism; for example, the information intensity increment of a successful role switch path increases by 10%.

[0062] The distributed consensus algorithm optimizes candidate execution paths, achieves path convergence through dynamic updates of information strength and path evaluation, and generates the optimal execution path (the minimum policy deformation path optimized by the distributed consensus, which satisfies the principle of maximizing group utility).

[0063] The distributed consensus algorithm can be implemented using the ant colony optimization algorithm, and the specific steps include: Step A1 Initialize candidate execution paths: Generate multiple candidate execution paths for each execution unit (such as resource allocation, task division). For example, candidate execution paths for resource allocation include "allocate budget evenly", "allocate budget by priority", and "allocate budget by demand".

[0064] Step A2 Define the path evaluation function: Define a global evaluation function to evaluate the execution effect of each execution path, such as v(P) = w_1×resource utilization + w_2×task completion rate + w_3×fairness, where w_1, w_2, w_3 are weight coefficients, w_1 = 0.4, w_2 = 0.3, w_3 = 0.3.

[0065] Step A3 Distributed Information Sharing: Each user group acts as a distributed node, maintaining its own local path evaluation results and exchanging information with other nodes through information sharing mechanisms (such as broadcasting and point-to-point communication). The shared content includes the local optimal path, path evaluation value, and group preferences (such as fairness of teacher group preferences).

[0066] Step A4 Path Optimization Iteration: Iteratively optimize the execution path using the ant colony optimization algorithm, including initializing information strength, path selection, path evaluation, and information strength updating.

[0067] Step A5 generates the optimal execution path: After the algorithm converges, the path with the highest information intensity is selected as the optimal execution path. If there are multiple high-intensity paths or divergences, the final path can be selected through voting mechanisms, negotiation mechanisms (such as reallocating resources), or enforcement mechanisms (such as superior intervention).

[0068] Based on the generated optimal execution path, the contribution of each user group to the optimal execution path is quantified through a contribution evaluation mechanism; The deviation detection mechanism detects differences in contribution, and when the difference exceeds a preset threshold, optimization suggestions are generated and task allocation is adjusted. The optimization suggestions are generated based on the dynamic role allocation of user groups. The path prediction model predicts the deviation probability of the adjusted task allocation. The input features of the path prediction model include the current game round, the strategies of each user group, the pheromone intensity, and the path evaluation value. The output is the path deviation probability of the future game round. When the predicted deviation probability exceeds a preset threshold, an intervention mechanism is triggered and the optimal execution path (optimal execution path) is regenerated. The intervention mechanism includes resource adjustment, strategy guidance, and negotiation mechanism. The feedback behavior of the group is collected and entropy reward is allocated through feedback closed-loop management. The feedback behavior includes reporting task allocation conflicts, confirming resource gaps, and proposing optimization solutions. The contribution, deviation prediction, and optimization suggestions are presented through visualization and multimodal interaction, including voice query, gesture adjustment, and haptic prompts.

[0069] The contribution evaluation mechanism can be implemented using the Shapley value to quantify the contribution of each user group in the optimal execution path. The formula for calculating the Shapley value is ϕ. i (v)=∑ S⊆N∖{i} (1 / ∣N∣( ∣N∣-1 ∣S∣ )×(v(S∪{i})−v(S)), where N is all user groups, S is a subset outside user group i, |S| is the size of subset S (i.e. the number of users it contains), v(S) is the performance effect of subset S (characteristic function value, representing the value of the subset alliance), and v(S∪{i})−v(S) is the marginal contribution of user i to subset S.

[0070] The deviation detection mechanism can be implemented by comparing the differences in Shapley values ​​among different user groups. For example, if the difference in Shapley values ​​between the teacher group and the principal group is greater than 0.7, it is determined to be an execution deviation. Optimization suggestions can be generated through a rule engine or machine learning model, combined with dynamic role allocation. For example, an optimization suggestion might be "increase the resource allocation ratio for the teacher group by 10% and switch teacher A to an administrative role to improve procurement efficiency."

[0071] The path prediction model can be implemented using a Long Short-Term Memory (LSTM) network or a Transformer model. For example, the prediction model might output, "The adjusted task assignment has an 80% probability of deviation in round 5 due to resource allocation conflict between the teacher group and the principal group." Training data can be generated from historical task assignment data.

[0072] Resource adjustment refers to automatically increasing the total amount of resources or reallocating resources, such as increasing the procurement budget by 10%. Strategy guidance refers to guiding user groups to choose low-bias paths through pheromone adjustments, such as increasing the pheromone intensity of low-bias paths by 20%. Negotiation mechanism refers to initiating virtual negotiations between user groups, such as prompting teacher groups and principal groups to negotiate resource allocation plans through a messaging system.

[0073] Through a closed-loop feedback management mechanism, user feedback is collected and entropy rewards are allocated. For example, reporting a task allocation conflict allocates 120 entropy rewards, confirming a resource shortage allocates 200 entropy rewards, and proposing and adopting an optimization plan allocates 500 entropy rewards. Resources are redeemed based on entropy rewards; for example, 1 entropy reward can be exchanged for 1 minute of automated report filling time, 200 entropy rewards for an accelerated equipment procurement channel, and 500 entropy rewards for regulatory exemptions. The allocation and redemption process of entropy rewards is recorded through a distributed ledger, and smart contracts automatically execute the redemption of entropy rewards and resource allocation. If feedback remains unprocessed, 10 entropy rewards are deducted from the regulator's account every second, and intervention from higher authorities is triggered when the entropy reaches zero.

[0074] The visualization is implemented using a 3D bar chart to show the distribution of Shapley values ​​for each user group, with deviation areas highlighted in red and optimization suggestions marked in green. Path prediction results are marked in advance with dashed boxes, for example, predicting possible deviation areas before task allocation adjustments. The path evolution process is displayed through a visual timeline, allowing users to view the contribution distribution at different time points by sliding the timeline. The multimodal interaction methods include: querying contribution via voice commands, such as automatically reading and highlighting relevant data after a user commands "show teacher group contribution"; adjusting task allocation via gestures, such as updating Shapley values ​​and providing feedback on optimization effects after a user drags the task allocation bar; and providing tactile feedback to alert users to deviations, such as providing high-frequency vibrations (5 consecutive short vibrations) when the deviation exceeds a threshold or the predicted deviation probability exceeds a threshold, allowing users to trigger automatic optimization by long-pressing the bracelet.

[0075] Regulators set up a feedback incentive pool and allocate incentive values ​​based on the feedback behavior of the regulated entities. Feedback behaviors include reporting policy contradictions (e.g., the regulated entity reports "uneven allocation of equipment procurement budget"), confirming resource gaps (e.g., the regulated entity confirms "less than 10 sets of classroom equipment"), and proposing optimization suggestions (e.g., the regulated entity proposes "an optimization plan to increase the procurement budget"). Resources are redeemed based on incentive values, which include automated task processing time, accelerated resource allocation channels, and regulatory exemption limits. The allocation and redemption process of incentive values ​​is recorded through a distributed recording mechanism, and resource allocation is achieved through an automated execution mechanism. The distributed recording mechanism can be implemented using blockchain technology, employing a distributed ledger to record the allocation and exchange of entropy values, ensuring transparency and immutability. The records can be formatted as key-value pairs, such as "User ID|Feedback Type|Entropy Value|Timestamp".

[0076] The automated execution mechanism can be implemented through smart contracts. For example, when the regulated party selects the accelerated channel for exchanging equipment, the smart contract automatically deducts 200 entropy value and triggers the procurement process.

[0077] Feedback status and incentive value balance are displayed through visualization and multimodal interaction. Multimodal interaction methods can include: submitting feedback via voice commands, such as automatically generating a feedback form and allocating entropy value rewards after the regulated individual issues the command "report policy contradictions"; exchanging resources via gestures, such as automatically deducting the corresponding entropy value after swiping right to select the "equipment procurement acceleration channel"; and reminding users of the feedback status via haptic feedback, such as a short vibration (200 milliseconds) provided by a smart bracelet when feedback is processed or resources are successfully exchanged. The dashboard can also display the regulated individual's entropy value balance and a list of exchangeable resources, with the feedback status displayed as a progress bar, green indicating processed and redeemable for timeout.

[0078] The feedback incentive pool is modeled as a virtual economic system used to incentivize those being regulated to participate in feedback. The incentive value uses entropy as the unit of measurement (currency). For example: reporting policy contradictions allocates 120 entropy points, confirming resource gaps allocates 200 entropy points, and proposing optimization suggestions and having them adopted allocates 500 entropy points.

[0079] 1 entropy point can be exchanged for 1 minute of automated task processing time (e.g., automated report filling), 200 entropy points for a resource allocation acceleration channel (e.g., an equipment procurement acceleration channel), and 500 entropy points for a regulatory exemption (e.g., exemption from one regulatory inspection). The resource exchange rules can be predefined through a rule engine or expert knowledge base. For example, the exchange ratio for automated task processing time can be dynamically adjusted according to the task complexity.

[0080] When feedback is not processed, the incentive value of the regulator is dynamically adjusted, and an intervention mechanism is triggered when the incentive value reaches a preset condition. The intervention mechanism includes notifying superiors and reassigning tasks. The specific implementation of dynamically adjusting the incentive value of the regulator can adopt a "negative feedback circuit breaker" mechanism. For example, when feedback is not processed, 10 units of entropy value are deducted from the regulator's value every second, and an intervention mechanism is triggered when the entropy value reaches zero. The specific implementation of the intervention mechanism includes notifying superiors (e.g., sending a reminder email to the superior regulatory department) and reassigning tasks (e.g., reassigning unprocessed feedback tasks to administrative roles through the dynamic role allocation mechanism in claim 4).

[0081] By using a distributed ledger to record feedback behavior, incentive value adjustments, and intervention processes, the records are formatted as key-value pairs; The path prediction model predicts the probability of deviation in feedback behavior. The input features of the path prediction model include the urgency of the feedback behavior, the scope of its impact, the historical processing time, and the pheromone intensity. The output is the probability of deviation in future feedback processing. The path prediction model can be implemented using a Long Short-Term Memory (LSTM) network or a Transformer model. Input features include the urgency of the feedback action (e.g., 90 for reporting a policy conflict), the scope of impact (e.g., 100 for a school-wide impact), historical processing time (e.g., an average processing time of 48 hours), and information strength (e.g., the information strength of the path or the information strength of the information). The output is the probability of deviation in future feedback processing. For example, the output could be "Feedback 1 has an 80% probability of processing deviation in the next 24 hours because of its high urgency but excessively long historical processing time." Training data can be generated from historical feedback data, such as recording the urgency, scope of impact, processing time, and final processing result of each feedback.

[0082] When the prediction deviation probability exceeds a preset threshold, optimization suggestions are generated and the feedback processing flow is adjusted. The optimization suggestions are generated based on the dynamic role allocation of user groups. This can be achieved through a rule engine or machine learning model, combined with the dynamic role allocation of user groups as described in claim 4. For example, an optimization suggestion might be "assign the processing task of feedback 1 to administrative role A to improve processing efficiency," where administrative role A is determined by the role switching rule in claim 4. When the prediction deviation probability exceeds a preset threshold (e.g., 70%), the feedback processing flow is adjusted, such as increasing the priority of feedback (e.g., raising the priority flag of feedback 1 from 80 to 95) or increasing processing resources (e.g., allocating additional automated task processing time).

[0083] Feedback tracking records, deviation predictions, and optimization suggestions are displayed through visualization and multimodal interaction methods, including voice queries, gesture operations, and tactile cues.

[0084] The visualization is implemented using a timeline chart to show feedback tracking records, with feedback actions represented by green nodes, incentive value adjustments by red nodes, and intervention processes by blue nodes. Path prediction results are marked in advance with dashed boxes, for example, predicting potential deviation areas before feedback processing. The evolution of the feedback processing flow is displayed through a visual timeline, allowing users to view the feedback status at different points in time by sliding the timeline. The multimodal interaction methods include: querying feedback records via voice commands, such as the system automatically reading the five most recent feedback records after the monitored person issues the command "query recent feedback"; filtering records via gestures, such as swiping left to view historical feedback and swiping right to view the intervention process; and providing tactile feedback to remind users of the feedback status, such as a short vibration (200 milliseconds) when a new feedback record is generated, and a high-frequency vibration (five consecutive short vibrations) when the probability of feedback processing deviation exceeds a threshold.

[0085] The key-value pair record format for feedback behavior is {User ID|Feedback Type|Processing Status|Timestamp}, the key-value pair record format for incentive value adjustment is {Regulator ID|Deduction Entropy Value|Triggering Condition|Timestamp}, and the key-value pair record format for intervention process is {Intervention Type|Intervention Target|Timestamp}. Example

[0086] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described method for information interaction in an educational system oriented towards educational supervision.

[0087] Since the electronic device described in this embodiment is an electronic device used to implement the information interaction method for educational systems oriented towards educational supervision objects according to the embodiments of this application, those skilled in the art can understand the specific implementation methods and various variations of the electronic device in this embodiment based on the information interaction method for educational systems oriented towards educational supervision objects described in the embodiments of this application. Therefore, how the electronic device implements the method in the embodiments of this application will not be described in detail here. Any electronic device used by those skilled in the art to implement the information interaction method for educational systems oriented towards educational supervision objects according to the embodiments of this application falls within the scope of protection of this application.

[0088] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0089] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for users of ordinary technical skills, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for information exchange in an education system oriented towards educational supervision targets, characterized in that, include: S1: The regulator sets up an information fusion pool, receives information from multiple information sources and distributes the information through a single transmission channel, and the regulated party uses multi-dimensional sensory interaction to distinguish, present and interact with the information. S2: The regulator sets up a pre-pool, dynamically updates the intensity and timeliness entropy of information, adjusts the decay rate of information intensity according to user behavior, and reminds the regulated by color differentiation and sensory interaction. S3: The regulator decomposes the information in a structured manner, defines the role pool and divides the user groups, sets up a game simulation pool to simulate the execution effect, generates candidate execution paths, updates the candidate execution paths through role switching rules, and generates the optimal execution path through dynamic updates of information intensity and path evaluation. S4: Regulators quantify user group contributions based on the optimal execution path, detect differences in contributions and generate optimization suggestions, predict deviation probabilities and trigger intervention mechanisms, collect feedback behaviors and allocate incentive values, record the allocation and redemption process of incentive values, allocate resources through automated execution mechanisms, dynamically adjust regulatory incentive values ​​and trigger intervention mechanisms, and optimize the feedback processing flow through path prediction models.

2. The information interaction method for an education system oriented towards educational supervision objects according to claim 1, characterized in that, S1 includes: Regulators set up an information fusion pool to receive information from multiple sources and add regulatory parameter clusters, which include the source identifier, priority identifier, and content summary of the information. The information is mapped to a uniform encoding format based on its source identifier and distinguished by the encoding value; then, the information is distributed through a single transmission channel. The regulatory parameter clusters for identifying information about the regulated entities are presented and interacted with using multi-dimensional sensory interaction. The regulator will transmit the information about adding regulatory parameter clusters to the front pool.

3. The information interaction method for an education system oriented towards educational supervision objects according to claim 2, characterized in that, S2 includes: For regulators, the initial intensity and initial time-limited entropy value of each information window are predefined according to the user role; A pre-pool is set up. After the information is identified and processed by the pre-pool, a uni-tripartite label (basic score | urgency, importance, timeliness) is added. The basic score is set according to urgency, importance, and timeliness. Identify the information type and fill it into the corresponding information window; Extract the average basic score of information within the window and dynamically update the intensity based on the information release frequency; extract the average urgency and importance of information within the window and dynamically update the updated time entropy value by combining the user's historical behavior and the initial time entropy value; customize the update cycle; The information intensity is calculated to decay naturally over time based on the time entropy value, and the decay rate is adjusted according to user behavior; the information intensity is updated according to the time frequency, and the information is displayed with different colors according to the threshold range it falls within.

4. The information interaction method for an education system oriented towards educational supervision objects according to claim 3, characterized in that, The method of updating information according to time frequency and differentiating colors based on the threshold range includes: If the intensity of the current information is higher than the threshold, it is recorded as high intensity information, highlighted with a high-saturation color (such as orange or red) and flashed at 1-second intervals, accompanied by a 500-millisecond prompt sound and 3 consecutive short vibrations. If the intensity of the current information is below the threshold, it is recorded as low intensity information. Low intensity information reduces its transparency with the information intensity update time frequency and fades out after the time limit is exceeded. Before fading out, a single long vibration is used to remind the user.

5. The information interaction method for an education system oriented towards educational supervision objects according to claim 4, characterized in that, S3 includes; The regulator decomposes the information issued to the regulated party into multiple execution units based on the initial intensity and time entropy value of the information, including resource allocation, task division and execution time limit; Define a pool of roles and assign specific functionalities, objective functions, and resource constraints to each role. Users in education supervision are divided into multiple groups, each group corresponds to a role, and each role has specific decision-making rules; Set up a game simulation pool to simulate the execution effects of different roles and generate multiple candidate execution paths; Predefined role switching rules are used to evaluate the probability of role allocation based on the user's historical characteristic data. When the probability is higher than expected, role switching is triggered and an allocation plan is generated. Candidate execution paths are updated based on the allocation plan. Path convergence is achieved through dynamic updates of information intensity and path evaluation, and the optimal execution path is generated. Regulators collect feedback from those they regulate in order to optimize task allocation and generate the optimal execution path.

6. The information interaction method for an education system oriented towards educational supervision objects according to claim 5, characterized in that, S4 includes: Based on the generated optimal execution path, the contribution of each user group is quantified through a contribution evaluation mechanism; The system detects differences in contribution levels. When the differences exceed a preset threshold, it generates optimization suggestions and adjusts task allocation. The optimization suggestions are generated based on the dynamic role allocation of user groups. Predict the probability of deviation in task allocation after adjustment. When the probability of deviation exceeds a preset threshold, trigger the intervention mechanism and regenerate the optimal execution path. The intervention mechanism includes resource adjustment, strategy guidance and negotiation mechanism. Collect feedback behavior from the group and allocate entropy rewards. Feedback behavior includes reporting task allocation conflicts, confirming resource gaps, and proposing optimization solutions.

7. The information interaction method for an education system oriented towards educational supervision objects according to claim 6, characterized in that, The collection of feedback behavior from the group and the allocation of entropy rewards include: Regulators set up a feedback incentive pool and allocate incentive values ​​based on the feedback behavior of the regulated entities. Feedback behavior includes reporting policy contradictions, confirming resource gaps, and proposing optimization suggestions. Resources are redeemed based on incentive values, which include automated task processing time, accelerated resource allocation channels, and regulatory exemption limits. Record the allocation and redemption process of incentive values, and realize resource allocation through an automated execution mechanism; The feedback status and incentive value balance are displayed through visualization and multimodal interaction.

8. The information interaction method for an education system oriented towards educational supervision objects according to claim 7, characterized in that, The feedback incentive pool is modeled as a virtual economic system, with the incentive value using entropy as the unit of measurement. 1 entropy value is exchanged for 1 minute of automated task processing time, 200 entropy values ​​are exchanged for a resource allocation acceleration channel, and 500 entropy values ​​are exchanged for a regulatory exemption quota.

9. A method for information interaction in an education system oriented towards educational supervision objects, as described in claim 8, characterized in that, When the feedback behavior is not processed, the incentive value of the regulator is dynamically adjusted, and an intervention mechanism is triggered when the incentive value reaches a preset condition. The intervention mechanism includes notifying superiors and reassigning tasks. By using a distributed ledger to record feedback behavior, incentive value adjustments, and intervention processes, the records are formatted as key-value pairs; The path prediction model predicts the probability of deviation in feedback behavior. The input features of the path prediction model include the urgency of the feedback behavior, the scope of its impact, the historical processing time, and the pheromone intensity. The output is the probability of deviation in future feedback processing. When the prediction deviation probability exceeds a preset threshold, optimization suggestions are generated and the feedback processing flow is adjusted.

10. A method for information interaction in an education system oriented towards educational supervision objects, as described in claim 9, characterized in that, The key-value pair record format for the feedback behavior is {User ID|Feedback Type|Processing Status|Timestamp}, the key-value pair record format for the incentive value adjustment is {Regulator ID|Deduction Entropy Value|Trigger Condition|Timestamp}, and the key-value pair record format for the intervention process is {Intervention Type|Intervention Object|Timestamp}.

Images