Digital education-oriented AI teaching assistance and teaching resource optimization and integration method

Abstract

The invention relates to the technical field of teaching management, and discloses a digital education-oriented AI teaching assistance and teaching resource optimization integration method, which comprises the following steps of: firstly, acquiring learning data of students, and defining a plurality of optimization targets; the method comprises the steps of establishing an optimization objective function based on a plurality of optimization objectives, solving the optimization objective function through a multi-objective optimization algorithm, dynamically adjusting a learning path according to real-time feedback of students by using a reinforcement learning algorithm based on student learning data, teaching resource use data and teacher work data which are collected in real time, and optimizing personalized learning experience. An Internet of Things sensor is used for collecting classroom environment data, and campus energy is intelligently managed in combination with an AI algorithm. According to the method and the system, multi-dimensional objective functions such as student learning progress and the like are constructed, and the Pareto optimal solution is solved by adopting the multi-objective optimization algorithm, so that collaborative scheduling of teaching resources under multiple objectives is realized, and the effect of greatly improving the resource allocation rationality of the whole teaching system is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of teaching management, and in particular to an AI teaching assistant and teaching resource optimization and integration method for digital education. Background Art

[0002] With the deep integration of information technology and the education industry, digital education has gradually become one of the mainstream teaching models. Forms such as online courses, intelligent learning platforms, virtual laboratories, and blended teaching have been continuously popularized, promoting the diversification of teaching resources and the personalization of learning methods. At the same time, the widespread application of artificial intelligence (AI) technology has also led to the emergence of intelligent education service forms such as "AI teaching assistants", which are used to assist teachers in managing teaching tasks, analyzing student behavior data, improving teaching efficiency, and the level of personalized services.

[0003] However, there are still many deficiencies in the existing digital education systems in terms of teaching resource management and intelligent regulation of learning paths. On the one hand, a large number of teaching resources have problems such as scattered distribution, duplicate construction, and low utilization rate. There is a lack of an effective unified scheduling mechanism, and it is difficult to maximize the resource utilization efficiency. On the other hand, although some platforms have introduced recommendation algorithms to assist students in course selection and resource matching, most are based on static rules or collaborative filtering methods, and it is difficult to dynamically respond to the real-time feedback and state changes of students during the learning process, resulting in insufficient personalized learning experiences.

[0004] In addition, teachers still need to undertake a large number of repetitive tasks in digital teaching (such as homework grading, answering questions and interactions, classroom management, etc.). Manual intervention is frequent, which is likely to cause an excessive teaching burden and is not conducive to teachers' focus on teaching content innovation. There are also common problems such as resource islands and course scheduling conflicts in the multi-campus deployment environment of teaching resources, and it is impossible to achieve efficient coordination of teaching supply and demand. At the same time, the current education system mostly ignores the energy consumption management and sustainable development needs in the campus environment, and lacks an intelligent control mechanism for the energy use of teaching equipment and a carbon emission feedback mechanism. Summary of the Invention

[0005] To make up for the above deficiencies, the present invention provides an AI teaching assistant and teaching resource optimization and integration method for digital education, aiming to improve the problem in the prior art that the scheduling of teaching resources usually only considers a single factor, such as course arrangement or resource usage frequency, lacking comprehensive consideration of multiple goals such as students' learning progress, teachers' burden, and resource utilization rate, resulting in unreasonable resource allocation and difficulty in meeting the actual needs of different users.

[0006] In the first aspect, the present invention provides the following technical solution. An AI teaching assistant and teaching resource optimization and integration method for digital education, which is applied to a server, includes the following steps: First, collect students' learning data and define multiple optimization goals; Based on multiple optimization objectives, an optimization objective function is established to quantify students' learning progress, resource utilization rate, and teachers' workload. The optimization objective function is solved through a multi-objective optimization algorithm to obtain multiple Pareto optimal solutions, balancing the relationships between different objectives; based on the real-time collected students' learning data, teaching resource usage data, and teachers' work data, the learning path, resource allocation plan, and teachers' workload are dynamically adjusted. The reinforcement learning algorithm is used to dynamically adjust the learning path according to students' real-time feedback, optimizing the personalized learning experience. IoT sensors are used to collect classroom environment data, and combined with AI algorithms, intelligent management of campus energy is carried out.

[0007] Preferably, the optimization objective function includes the following sub-steps: Step 1: The objective function for maximizing students' learning progress, which evaluates students' progress in different learning paths and is used to maximize students' learning. Step 2: The objective function for maximizing resource utilization, which evaluates the demand, usage frequency, and priority of various teaching resources and is used to maximize resource utilization efficiency. Step 3: The objective function for minimizing teachers' workload, which optimizes the teachers' task allocation plan, reduces the non-teaching workload of teachers, and enables teachers to focus on teaching content.

[0008] Preferably, the objective function for minimizing teachers' workload is used to assist teachers in correcting homework, answering questions, and classroom management tasks through an AI teaching assistant system, thereby reducing the non-teaching workload of teachers and helping teachers focus on core teaching content. The optimization objectives include maximizing students' learning progress, maximizing resource utilization, and minimizing teachers' workload. The multi-objective optimization algorithm adopts genetic algorithms, particle swarm optimization algorithms, or other evolutionary algorithms to search for multiple Pareto optimal solution sets and obtain the optimal solution for balancing each objective.

[0009] Preferably, the real-time data includes students' learning behavior data, resource usage data, and teachers' workload data. The data is collected in real-time through intelligent sensors and the learning management system and is input into the optimization algorithm for real-time analysis and dynamic adjustment.

[0010] Preferably, the reinforcement learning algorithm adopts the Q-learning algorithm. The Q-learning algorithm adjusts the learning path and recommended teaching resources through students' real-time feedback, for personalized learning plans, maximizing learning education.

[0011] Preferably, the Internet of Things sensors are used to collect environmental data on campus, including temperature, humidity, light intensity, and equipment usage. The environmental data is used to adjust the working states of air conditioners and lighting equipment in classrooms. The teaching resources include course videos, teaching question banks, laboratory equipment usage rights, and e - textbooks. The resource utilization maximization objective function dynamically evaluates and optimizes resource allocation by assessing the usage frequency, demand, and priority of resources.

[0012] Preferably, the method further includes: Setting up a carbon footprint calculation module, which is used to generate a campus energy - saving report and optimize energy usage according to the report data to reduce unnecessary energy consumption and carbon emissions; Based on the teaching resource requirements and utilization situations of multiple campuses, designing a cross - campus resource sharing strategy to dynamically adjust the teaching resource allocation between different campuses through an optimization algorithm; Adopting an intelligent class scheduling strategy to avoid resource redundancy, overlap between campuses, and improve the utilization efficiency of teaching resources.

[0013] In a second aspect, the present invention provides the following technical solution: an AI teaching assistant and teaching resource optimization and integration system for digital education. The system includes: An optimization objective modeling module, which is used to define multiple optimization objectives based on students' learning progress and establish an objective function for quantifying students' learning progress, resource utilization rate, and teachers' workload; A multi - objective optimization calculation module, which is used to solve the objective function through a multi - objective optimization algorithm to obtain multiple Pareto optimal solutions; a data collection and perception module, which is used to collect students' learning behavior data, teaching resource usage data, teachers' workload data, and campus environmental data, and input the behavior data into the optimization algorithm for dynamic adjustment; A personalized learning regulation module, which is used to process students' real - time feedback based on a reinforcement learning algorithm to optimize the personalized learning experience; an energy management control module, which is used to intelligently control classroom environmental equipment based on the environmental data collected by Internet of Things sensors; a resource sharing and scheduling module, which is used for cross - campus teaching resource sharing strategies, dynamically adjusts the resource allocation plan according to resource usage situations and campus requirements, and improves the overall resource utilization rate through an intelligent class scheduling strategy; A carbon emission assessment and energy - saving feedback module, which is used to calculate and output a campus carbon footprint report and analyze the energy consumption and emission situations.

[0014] In a third aspect, the invention provides the following technical solution: a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the above - mentioned method for optimizing and integrating an AI teaching assistant and teaching resources for digital education.

[0015] Fourth aspect, the present invention provides the following technical solution: a readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the above-mentioned method for optimizing the integration of an AI teaching assistant and teaching resources for digital education.

[0016] The present invention has the following beneficial effects: 1. In the present invention, by constructing multi-dimensional objective functions such as students' learning progress, teachers' burden, and resource utilization rate, and using a multi-objective optimization algorithm to solve the Pareto optimal solution, the collaborative scheduling of teaching resources under multiple objectives is realized, and the rationality of the overall teaching system resource configuration is greatly improved.

[0017] 2. In the present invention, by introducing the reinforcement learning algorithm Q-learning, the students' learning behavior feedback is processed in real time and the learning path and resource recommendation are dynamically adjusted, so as to realize the intelligent control of the individualized learning experience, and the learning efficiency and adaptability are significantly improved.

[0018] 3. In the present invention, by designing a cross-campus resource sharing strategy and an intelligent course scheduling mechanism, and dynamically scheduling teaching resources in combination with the resource usage status and differences between campuses, the effective avoidance of resource duplication configuration is realized, and the overall resource utilization rate of the education system is improved and teaching conflicts are reduced.

[0019] 4. In the present invention, by deploying Internet of Things sensors and a carbon footprint calculation module, the campus environment and equipment energy consumption data are collected in real time and an energy-saving report is generated, so as to realize the intelligent control of classroom environment equipment and the optimization of energy consumption strategies, and the energy-saving and environmental protection effects of reducing energy waste and carbon emissions are obtained. Description of the Drawings

[0020] Figure 1 is a method flow chart of a method for optimizing the integration of an AI teaching assistant and teaching resources for digital education proposed by the present invention; Figure 2 is a module relationship architecture diagram of a system for optimizing the integration of an AI teaching assistant and teaching resources for digital education proposed by the present invention. Detailed Embodiments

[0021] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0022] Embodiment 1 Refer to Figure 1, in the first embodiment of the present invention, the present invention provides an AI teaching assistant and teaching resource optimization integration method for digital education, which is applied to a server and includes the following steps: First, collect students' learning data and define multiple optimization goals; Based on multiple optimization goals, establish an optimization objective function to quantify students' learning progress, resource utilization rate, and teachers' workload; Solve the optimization objective function through a multi-objective optimization algorithm to obtain multiple Pareto optimal solutions and balance the relationships between different objectives; based on the real-time collected students' learning data, teaching resource usage data, and teachers' work data, dynamically adjust the learning path, resource allocation plan, and teachers' workload; Use the reinforcement learning algorithm to dynamically adjust the learning path according to students' real-time feedback and optimize the personalized learning experience; Use Internet of Things sensors to collect classroom environment data and combine AI algorithms to intelligently manage campus energy.

[0023] The optimization objective function includes the following sub-steps: Step 1, the objective function for maximizing students' learning progress. The objective function is used to maximize students' learning by evaluating students' progress in different learning paths; Step 2, the objective function for maximizing resource utilization. The objective function is used to maximize the resource utilization efficiency by evaluating the demand, usage frequency, and priority of various teaching resources; Step 3, the objective function for minimizing teachers' workload. The objective function is used to enable teachers to focus on teaching content by optimizing the task allocation plan for teachers and reducing their non-teaching workload.

[0024] Specifically, this method runs in a server environment, relying on the large-scale data stream in the education scenario and combining intelligent optimization algorithms, aiming to dynamically construct and adjust students' personalized learning paths, improve the utilization efficiency of teaching resources, and reduce the repetitive labor pressure on teachers, so as to realize the collaborative intelligent management of the teaching process. By using a multi-objective optimization strategy to process multiple core indicators, specifically including three dimensions: students' learning progress, teaching resource allocation efficiency, and teachers' task burden. In actual deployment, these objectives often have problems of mutual restriction. For example, a curriculum arrangement with highly concentrated resources may improve learning efficiency, but it may also cause uneven teacher loads. Therefore, this method introduces the Pareto multi-objective optimization framework to construct multiple optimal solution sets for the strategy scheduling module to make decision selections, ensuring the balance and flexibility of the overall teaching system operation; Step 1, establish an objective function for maximizing students' learning progress. This objective function is used to measure the actual learning progress of students under different learning paths. Usually, the number of knowledge points completed by students per unit time, task completion rate, or understanding degree evaluation value is used as the basic variable, The following mathematical forms can be introduced for modeling.

[0025] Among them, f1(x) represents the objective function of the student's learning progress; x is the feature vector of the learning path where the current student is located; n is the total number of learning tasks; P i (x) represents the completion degree of the student in the i-th task, and the value range is from 0 to 1; α i is the weight of the task, which is set according to the course difficulty and the key degree of knowledge points.

[0026] The weight α i can be determined through historical data clustering analysis or set according to the experience of course designers. The P value can be derived from the completion records, error rates, or learning time indicators in the learning platform. In actual deployment, it can be collected in real time in combination with the interface of the learning management system (LMS); to enhance the adaptability to individual differences, a student ability coefficient or a cognitive curve parameter can be introduced to perform personalized modulation on f1(x); Step 2: Construct the objective function for maximizing resource utilization Resource utilization rate is one of the important indicators to measure the operation efficiency of the teaching system. Its core lies in evaluating the usage frequency, coverage range, and urgency of various teaching resources, and avoiding resource redundancy, idleness, or repeated delivery. The function for maximizing resource utilization can be expressed as: Among them: f2(x) represents the objective function of resource utilization efficiency; m represents the number of types of teaching resources; U j (x) represents the usage frequency or call times of the j-th type of resource; D j is the current supply of this type of resource or the upper limit that the system can allocate; β j is the resource urgency / priority coefficient, which is preset according to the strategy requirements.

[0027] Resource types include course videos, online question banks, virtual experiment resources, teacher Q&A time windows, etc. The system docks with the LMS and CMS interfaces through the resource scheduling module to dynamically update the value of U j To improve the accuracy of resource scheduling, a resource heat change factor can be introduced to establish a resource aging model to reduce the priority of resources that have not been used for a long time; Step 3: Design the objective function for minimizing the teacher's workload: As the core of the teaching process, the burden on teachers directly affects the quality of education and the efficiency of service response. Through a task allocation optimization mechanism, this invention reduces the participation of teachers in repetitive work and improves their concentration in the teaching process. The form of this objective function is as follows: Among them, f3(x) represents the objective function of the teacher's burden, and taking the negative sign represents minimizing the burden; q represents the total number of types of teachers' tasks; T k (x) is the evaluation value of the time consumed or complexity of the kth type of task; γ k is the weight of task importance / irreplaceability; Task types can include homework grading, answering questions, content assignment, classroom interaction, resource preparation, etc. In a possible implementation, the AI teaching assistant module undertakes some tasks, such as automatically grading objective questions, semantically recognizing and answering common questions, and recommending course resources, thereby greatly reducing several dimensional values in T k (x). A working time constraint function can also be introduced to set soft and hard limits on the total daily burden of teachers, and combined with the course scheduling engine to achieve the balance of teacher task allocation; Based on the above objective function, an overall optimization model is constructed: Maximize F(x)=(f1(x),f2(x),f3(x)); A multi-objective optimization algorithm is used for solving. Common choices include: NSGA-II, MOEA / D, evolutionary search based on reinforcement learning, etc. An initial population is constructed through a genetic algorithm, the objective function is scored for each individual, and through crossover and mutation operations, continuous iterative optimization is carried out to finally obtain a set of Pareto optimal solutions that meet different preferences. The scheduler selects the appropriate solution set according to the current policy weight, resource status and teaching objectives to achieve dynamic teaching control; In the application process, a real-time feedback mechanism is also combined, and the Q-learning algorithm is introduced for personalized policy update. The system uses the student feedback score, error rate, and learning delay as state variables, and the teaching adjustment result as the action set, and drives the policy adjustment through the reward function, so as to achieve the reinforcement adaptability of the learning path. In the application, the system is deployed on a cloud server or an edge computing node, supports multi-campus and multi-grade operation at the same time, and the teaching data can be integrated with the existing educational administration platform through the API, with good engineering adaptability and scalability.

[0028] The objective function for minimizing teachers' workload is used to assist teachers in correcting homework, answering questions, and managing classroom tasks through an AI teaching assistant system, thereby reducing teachers' non-teaching workload and helping teachers focus on core teaching content. The optimization objectives include maximizing students' learning progress, maximizing resource utilization, and minimizing teachers' workload. The multi-objective optimization algorithm uses genetic algorithms, particle swarm optimization algorithms, or other evolutionary algorithms to search for multiple Pareto optimal solution sets and obtain the optimal solution that balances between various objectives.

[0029] Specifically, through the multi-objective optimization strategy, learning efficiency is improved, resource allocation is optimized, and teachers' workload is reduced. Processes such as establishing objective functions, multi-objective optimization solving, and dynamic feedback adjustment are established. Multiple optimization objectives are balanced simultaneously through algorithms to improve the intelligent level of the overall teaching system. The objective function for maximizing students' learning progress is used to measure the effect of the learning path, and the objective function is defined as: where P i (x) is the task completion rate, α i is the weight, and the weight can be set according to the importance of knowledge points or course requirements. The P value is generated by the system collecting students' behavior data, and a student ability parameter can also be introduced to adjust the model sensitivity. The objective function for maximizing resource utilization reflects the teaching resource scheduling efficiency and is defined as follows: where U j (x) is the number of times the resource is used, D j is the resource supply, β j represents the priority. The system collects resource usage information in real time and dynamically adjusts the scheduling strategy. The objective function for minimizing teachers' workload is assisted by an AI teaching assistant to complete non-teaching tasks, and the objective function is extended to: where A k (x) is the amount of tasks automatically completed by the AI. The AI module can correct homework, provide automatic answers to questions, and manage classroom behavior, reducing teachers' repetitive work. In the application, the AI is embedded through the LMS system to replace routine tasks, and teachers can focus on content teaching and personalized tutoring.

[0030] The multi-objective optimization algorithm uses evolutionary algorithms such as genetic algorithms and particle swarm optimization to search for multiple Pareto optimal solutions, and each solution represents a set of teaching strategy combinations: F(x) = (f1(x), f2(x), f3(x)); The NSGA-II algorithm is used to complete the non-dominated sorting of solutions, obtaining a scheduling combination with target balance. In some cases, PSO and GA can be used in combination to improve the search efficiency. The finally output policy plan is selected from the solution set according to the teaching priority or real-time demand to adapt to the scenarios of different schools or teaching stages.

[0031] Real-time data includes students' learning behavior data, resource usage data, and teachers' workload data. The data is collected in real time through intelligent sensors and the learning management system and input into the optimization algorithm for real-time analysis and dynamic adjustment.

[0032] Specifically, the dynamics of the optimization algorithm rely on the system's continuous acquisition and analysis of multi-source real-time data. To ensure the timeliness and adaptability of the optimization results, the system introduces a real-time data collection mechanism to obtain key behavior data in the teaching environment through various sensing means and platform interfaces; Real-time data mainly includes the following three categories: Students' learning behavior data, such as learning duration, task completion status, answer correctness rate, frequency of learning path jumps, etc.; teaching resource usage data, including video resource call frequency, question bank access times, virtual experiment platform usage, etc.; Teachers' workload data, such as the amount of marking tasks, number of answering questions, platform interaction duration, frequency of classroom management interventions, etc.; students' behavior data can be automatically collected through the learning management system (LMS) or classroom interaction system. The platform records the stay time, operation trajectory, and completion status of each student in different resource modules and conducts structured sorting through the built-in log system; resource usage data usually comes from the content management system (CMS) and teaching service middleware, such as cloud classrooms, homework systems, experiment platforms, etc. The system can call its open API to extract the current usage status and call frequency of various resources in real time; Regarding teachers' workload data, it can be collected through the teaching assistant tool on the teacher side or combined with forms such as time statistics plugins, task assignment panels, and voice interaction logs to comprehensively evaluate teachers' work intensity and task density; Classroom cameras can assist in judging the distribution of students' attention, and infrared attendance devices are used to collect attendance. Learning behaviors on tablets / PCs are recorded by the front end and transmitted to the back end to build a complete real-time status data set; All collected data is preliminarily filtered and preprocessed by edge computing nodes. After removing abnormal and redundant values, it is uniformly input into the optimization engine. The system dynamically adjusts the policy weights and optimization directions according to the current data status to achieve immediate optimization of students' path recommendations, resource allocation, and teachers' task arrangements. Secondly, a local recalculation mechanism with a short period (such as 30 minutes or 1 hour) is set. If a significant change in students' status or resource bottlenecks is detected, the re-optimization process is automatically triggered to adapt to temporary changes in the teaching scenario.

[0033] The reinforcement learning algorithm adopts the Q-learning algorithm, which adjusts the learning path and recommended teaching resources through the real-time feedback of students for personalized learning plans to maximize learning and education.

[0034] Specifically, the policy update of Q-learning is based on the following formula: As a model-free reinforcement learning method, Q-learning can learn the optimal policy through interaction without relying on an environmental model. The system regards each student as a learning agent, and their behavioral states and feedback results constitute the basic information for adjusting the learning path.

[0035] Among them: Q(s,a): represents the expected reward for taking action a in state s; α: learning rate, used to control the update amplitude of the Q value; r: the actual obtained reward value; γ: discount factor, indicating the importance of future rewards; s ′ : the next state after the execution of action a; a ′ : the next possible action in state s ′ .

[0036] In a typical implementation, the system defines that: The state space consists of indicators such as the student's current knowledge mastery, learning fatigue, wrong-question density, and completion progress. The action space includes switching learning paths, changing the recommended resource type, adjusting the task difficulty level, etc. The reward function is composed of multiple dimensions such as student feedback scores, task completion time, and learning satisfaction. After the student completes a certain task, the system calculates the immediate reward based on task duration, correct rate, click behavior, etc., and uses it to update the Q-value table. The system selects subsequent learning actions according to the principle of maximizing the Q value to achieve path strategy adaptation. To avoid drastic fluctuations in the learning path during the initial exploration period, an ε-greedy strategy can be introduced to balance the exploration and exploitation processes, or a dynamic learning rate mechanism can be introduced to improve the convergence speed; At the deployment level, the Q-learning module can be embedded in the teaching recommendation engine and connected to the LMS system interface to achieve individual-level path planning updates. For example, if a student continuously completes multiple tasks and performs excellently, the system can dynamically adjust their subsequent learning path to a higher level; if a student continuously has a high wrong-question rate, the system switches to more auxiliary resources or a repeated task module.

[0037] IoT sensors are used to collect environmental data on campus, including temperature, humidity, light intensity, and equipment usage. The environmental data is used to adjust the working states of air conditioners and lighting equipment in classrooms. Teaching resources include course videos, teaching question banks, the right to use laboratory equipment, and e-textbooks. The objective function of maximizing resource utilization dynamically evaluates and optimizes resource allocation by assessing the usage frequency, demand, and priority of resources.

[0038] Specifically, to improve the intelligent level of the teaching environment, the present invention introduces an IoT sensing mechanism into the system. Environmental state parameters in classrooms and on campus are collected through distributed sensors to drive the operation of intelligent devices, realizing the coordinated control of energy and teaching comfort. The environmental data includes temperature, humidity, light intensity, and equipment usage. Environmental sensors are deployed on classroom ceilings, by windows, at equipment interfaces, or at network gateway nodes, with the ability to sample in real time and connect to the system's main control platform wirelessly, uploading to the backend environmental control module. The system dynamically adjusts the operation states of devices according to the collected environmental data. For example, when the room temperature exceeds the set range, the air conditioner temperature is automatically adjusted; when the light intensity is insufficient, the light brightness is automatically started or adjusted; when no one is detected, the multimedia device is turned off to avoid idling. The teaching resources involved are rich in variety, supporting the unified scheduling and intelligent recommendation of multiple types of resources. The teaching resources mainly include course video resources (classified by subject, chapter, and difficulty); teaching question banks (used for homework pushing and exam generation); virtual laboratory equipment permissions; e-textbooks and extended reading materials. The resources are organized as structured objects, with unique identifiers and attribute tags, supporting being scheduled, allocated, recommended, and recycled by the system. The resource status is recorded in real time, including usage frequency, number of active users, remaining licenses, etc. The objective function of maximizing resource utilization is dynamically optimized based on the usage frequency, demand, and priority of resources. During the scheduling process, the system combines real-time data and resource status to hierarchically control the resources. Some resources adopt a shared pool mechanism, flowing between multiple campuses or different classes as needed to ensure the maximization of resource utilization efficiency and avoid redundancy and waste.

[0039] The method further includes: Setting up a carbon footprint calculation module to generate a campus energy-saving report and optimize energy usage according to the report data to reduce unnecessary energy consumption and carbon emissions; Designing a cross-campus resource sharing strategy based on the teaching resource requirements and utilization situations of multiple campuses, and dynamically adjusting the teaching resource configuration between different campuses through an optimization algorithm; Adopting an intelligent course scheduling strategy to avoid resource redundancy, overlap, and improve the utilization efficiency of teaching resources between campuses.

[0040] Specifically, the system is equipped with a carbon footprint calculation module that combines environmental and device data collected by IoT sensors to calculate the carbon emissions of various energy consumptions. This includes: the usage duration and power consumption of devices such as air conditioners, lights, and multimedia; the average energy consumption density per unit time in each teaching venue; peak / low peak energy consumption period analysis. The system generates a carbon emission assessment report based on the energy consumption structure and emission factors, provides detailed data display by building, classroom, and time dimensions, shuts down some devices during some non-teaching peak periods, and delays turning on lights in areas with sufficient light, reducing unnecessary energy consumption through scenario zoning and strategy setting; To improve the system utilization rate of the overall teaching resources, the system supports a cross-campus resource sharing and scheduling mechanism. In the scenario of multi-campus deployment, there are spatio-temporal mismatch problems in the resource supply and demand of different teaching points. The present invention models the resource usage situations of each campus through a scheduling algorithm, evaluates the sharing feasibility based on indicators such as usage frequency, demand intensity, and resource allocatability. The system distributes teaching resources to each campus as needed through unified resource identification and establishing a shared scheduling pool, including course content permissions, virtual experiment environments, teacher online guidance time, etc. When the resource load of a certain campus is too high or the usage efficiency is too low, the system can automatically adjust the resource allocation strategy and allocate redundant resources to high-demand campuses without affecting local tasks; An intelligent course scheduling mechanism is used to avoid duplicate arrangements or waste of teaching resources among multiple campuses. The system comprehensively establishes a multi-dimensional scheduling model based on course types, teacher arrangements, classroom usage plans, and students' learning paths. It uses a hybrid method of a rule engine and an optimization algorithm to intelligently generate a teaching schedule. During the scheduling, the system can identify typical problems such as resource overlap, device conflicts, and teacher time conflicts, and automatically adjust and avoid conflicts through algorithms. For example, if the laboratory equipment in a certain campus is temporarily unavailable, the system can adjust the corresponding course to a campus that supports remote experiment functions or other time periods, ensuring the smooth progress of the course while avoiding resource waste. The course scheduling result will be linked to the real-time resource status to achieve a linkage closed-loop of task adjustment and resource scheduling.

[0041] Embodiment 2: Refer to Figure 2 , in the second embodiment of the present invention, the present invention provides an AI teaching assistant and teaching resource optimization and integration system for digital education. The system includes: An optimization objective modeling module is used to define multiple optimization objectives based on students' learning progress and establish objective functions for quantifying students' learning progress, resource utilization rate, and teachers' work burden; The multi-objective optimization calculation module is used to solve the objective function through the multi-objective optimization algorithm to obtain multiple Pareto optimal solutions; the data collection and perception module is used to collect students' learning behavior data, teaching resource usage data, teachers' workload data, and campus environment data, and input the behavior data into the optimization algorithm for dynamic adjustment; The personalized learning regulation module is used to process students' real-time feedback based on the reinforcement learning algorithm to optimize the personalized learning experience; the energy management control module is used to intelligently control the classroom environment equipment based on the environmental data collected by the Internet of Things sensors; the resource sharing and scheduling module is used for the cross-campus teaching resource sharing strategy, dynamically adjusts the resource allocation plan according to the resource usage situation and campus needs, and improves the overall utilization rate of resources through the intelligent course scheduling strategy; The carbon emission assessment and energy-saving feedback module is used to calculate and output the campus carbon footprint report and analyze the energy consumption and emission situation.

[0042] Specifically, the optimization objective modeling module is used to construct multiple optimization objective functions based on students' learning progress, course plans, and teachers' teaching arrangements. The objectives include maximizing the adaptability of students' learning paths, maximizing the utilization rate of teaching resources, and minimizing teachers' workload. The module conducts quantitative modeling through learning behavior indicators (such as task completion rate, error rate, click frequency, etc.), resource status indicators (such as access frequency, concurrent user number, idle time period), and teachers' work intensity indicators (such as teaching duration, course concurrency number) to form a set of objective functions that can be called by the multi-objective optimization algorithm; the multi-objective optimization calculation module uses multi-objective evolutionary algorithms such as NSGA-II to solve the above objective functions, generates a set of representative Pareto optimal solutions, and is used to find a balance solution among multiple conflicting objectives, while taking into account students' personalized learning needs, the rationality of teachers' course scheduling, and the efficiency of teaching resource allocation. The optimization results can drive subsequent learning scheduling and teaching plan reconstruction; The data collection and perception module is deployed on the client, teaching terminal, and campus equipment, and collects students' learning behavior data (click behavior, task completion degree, feedback evaluation, etc.), teaching resource usage data (course access volume, question bank call times, experimental resource allocation records, etc.), teachers' workload data (teaching schedule, correction tasks, online interaction duration, etc.), and campus environment data (temperature, humidity, light, equipment operation status, etc.) through the Internet of Things technology and platform interface. After the data is collected, it is uniformly transmitted to the central scheduling engine for the optimization calculation module to call, and is used as the real-time input for intelligent control and personalized learning adjustment; Personalized learning regulation module It is used to process the real-time learning feedback data of students based on the Q-learning reinforcement learning algorithm, including learning progress, answering accuracy rate, learning stagnation time, etc. The system models each student as an independent learning entity, and through state transition and reward mechanisms, continuously updates the optimal learning path and teaching resource recommendation strategy. The Q-value update process conforms to the following formula: where s represents the current state of the student, a represents the recommended or task selection behavior, r is the feedback reward, s ′ is the next state, α is the learning rate, and γ is the discount factor. Through continuous feedback loops, the system realizes the regulation of highly personalized learning experiences.

[0043] An energy management control module that, by accessing the Internet of Things sensors deployed in teaching buildings, can obtain real-time classroom environment parameters (including temperature, humidity, light intensity, etc.) and equipment operation data (including the status of air conditioners, lighting, multimedia devices, etc.). The system intelligently adjusts the equipment operation strategy based on the environment and class schedule. For example, it automatically delays the turning on of lights when there is sufficient light and automatically turns off equipment when the classroom is idle, improving energy usage efficiency and user comfort; A resource sharing and scheduling module that, for a multi-campus or multi-classroom environment, constructs a unified teaching resource pool. The resources include course videos, experimental equipment, question banks, teacher guidance time, etc. The module quantitatively evaluates the usage frequency, access concurrency, and current demand of resources in each campus, and dynamically adjusts the resource allocation ratio of each campus in combination with optimization algorithms to achieve cross-regional resource sharing and avoid redundant configurations. The module also integrates an intelligent class scheduling strategy, which automatically generates the optimal class schedule based on course priorities, resource status, teacher time windows, etc., to avoid course conflicts and resource idleness.

[0044] A carbon emission assessment and energy-saving feedback module that, based on the energy consumption data and equipment operation logs collected by the energy management control module, calculates the total carbon emissions by equipment category and operation time period, and generates a carbon footprint analysis report. The report content includes an emission trend curve, an energy consumption structure distribution map, and energy-saving suggestions. The system can optimize the equipment operation strategy based on the report content, and early warning of high-carbon consumption areas or abnormal energy consumption behaviors, thereby further optimizing the campus energy structure and reducing the overall carbon emission level.

[0045] Embodiment III In the third embodiment of the present invention, based on the same inventive concept, a computer-readable storage medium proposed by the present invention stores a computer program, and when the computer program is executed by a processor, it realizes the steps of an AI teaching assistant and teaching resource optimization and integration method for digital education in the above embodiment.

[0046] Embodiment IV According to a fourth embodiment of the present invention, based on the same inventive concept, a computer device proposed by the present invention includes: a processor and a memory; the processor and the memory communicate with each other; the memory is used to store instructions; the processor is used to execute the instructions in the memory to execute the method for optimizing the integration of an AI teaching assistant and teaching resources for digital education in the above embodiment.

[0047] It should be understood that each part of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0048] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An AI teaching assistant and teaching resource optimization and integration method for digital education, applied to a server, characterized in that Including the following steps: First, collect students' learning data and define multiple optimization objectives; Based on multiple optimization objectives, establish an optimization objective function to quantify students' learning progress, resource utilization rate, and teachers' workload; Solve the optimization objective function through a multi-objective optimization algorithm to obtain multiple Pareto optimal solutions and balance the relationships between different objectives; Based on the real-time collected students' learning data, teaching resource usage data, and teachers' work data, dynamically adjust the learning path, resource allocation plan, and teachers' workload; Use a reinforcement learning algorithm to dynamically adjust the learning path according to students' real-time feedback and optimize the personalized learning experience; Use Internet of Things sensors to collect classroom environment data and combine with AI algorithms to intelligently manage campus energy.

2. The AI teaching assistant and teaching resource optimization and integration method for digital education according to claim 1, wherein, The optimization objective function includes the following sub-steps: Step 1: The objective function for maximizing students' learning progress. This objective function is used to maximize students' learning by evaluating students' progress in different learning paths; Step 2: The objective function for maximizing resource utilization. This objective function is used to maximize resource utilization efficiency by evaluating the demand, usage frequency, and priority of various teaching resources; Step 3: The objective function for minimizing teachers' workload. This objective function is used to enable teachers to focus on teaching content by optimizing teachers' task allocation plans and reducing teachers' non-teaching workload; 3. An AI teaching assistant and teaching resource optimization and integration method for digital education according to claim 1, characterized in that The objective function for minimizing teachers' workload is used to assist teachers in correcting homework, answering questions, and classroom management tasks through an AI teaching assistant system, thereby reducing teachers' non-teaching workload and helping teachers focus on core teaching content. The optimization objectives include maximizing students' learning progress, maximizing resource utilization, and minimizing teachers' workload; The multi-objective optimization algorithm uses a genetic algorithm, particle swarm optimization algorithm, or other evolutionary algorithms to search for multiple Pareto optimal solution sets and obtain the optimal solution for balancing each objective; 4. An AI teaching assistant and teaching resource optimization and integration method for digital education according to claim 1, characterized in that, The real-time data includes students' learning behavior data, resource usage data, and teachers' workload data. The data is collected in real time through intelligent sensors and the learning management system and is input into the optimization algorithm for real-time analysis and dynamic adjustment; 5. A method for optimizing the integration of an AI teaching assistant and teaching resources for digital education according to claim 1, characterized in that, The reinforcement learning algorithm uses the Q-learning algorithm. The Q-learning algorithm adjusts the learning path and recommended teaching resources through students' real-time feedback for personalized learning plans and maximizes learning education; 6. The AI teaching assistant and teaching resource optimization integration method for digital education according to claim 1, characterized in that The Internet of Things sensors are used to collect environmental data on campus, including temperature, humidity, light intensity, and equipment usage. The environmental data is used to adjust the working status of air conditioners and lighting equipment in the classroom. The teaching resources include course videos, teaching question banks, laboratory equipment usage rights, and e-books. The objective function for maximizing resource utilization dynamically evaluates and optimizes resource allocation by evaluating the usage frequency, demand, and priority of resources; 7. An AI teaching assistant and teaching resource optimization and integration method for digital education according to claim 1, characterized in that, The method further includes: Set up a carbon footprint calculation module to generate a campus energy-saving report and optimize energy usage according to the report data to reduce unnecessary energy consumption and carbon emissions; Based on the teaching resource requirements and utilization situations of multiple campuses, design cross-campus resource sharing strategies, and dynamically adjust the teaching resource allocation between different campuses through an optimization algorithm; Through an intelligent course scheduling strategy, it is used to avoid resource redundancy, overlap, and the utilization efficiency of teaching resources between campuses.

8. An AI teaching assistant and teaching resource optimization and integration system for digital education, characterized in that, For an AI teaching assistant and teaching resource optimization integration method for digital education according to any one of claims 1-7, the system includes: An optimization objective modeling module, which is used to define multiple optimization objectives based on the learning progress of students, and establish an objective function for quantifying the learning progress of students, resource utilization rate, and teachers' workload; A multi-objective optimization calculation module, which is used to solve the objective function through a multi-objective optimization algorithm to obtain multiple Pareto optimal solutions; A data collection and perception module, which is used to collect students' learning behavior data, teaching resource usage data, teachers' workload data, and campus environment data, and input the behavior data into the optimization algorithm for dynamic adjustment; A personalized learning regulation module, which is used to process students' real-time feedback based on a reinforcement learning algorithm to optimize the personalized learning experience; An energy management control module, which is used to intelligently control classroom environment equipment based on the environmental data collected by Internet of Things sensors; A resource sharing and scheduling module, which is used for cross-campus teaching resource sharing strategies, dynamically adjusts the resource allocation plan according to the resource usage situation and campus requirements, and improves the overall resource utilization rate through an intelligent course scheduling strategy; A carbon emission assessment and energy-saving feedback module, which is used to calculate and output a campus carbon footprint report, and analyze the energy consumption and emission situations.

9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it realizes an AI teaching assistant and teaching resource optimization integration method for digital education according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that, A computer program is stored on the readable storage medium, and when the computer program is executed by the processor, it realizes an AI teaching assistant and teaching resource optimization integration method for digital education according to any one of claims 1 to 7.

Images