A comprehensive management system for smart digital campus

Abstract

The present invention discloses a comprehensive management system for a smart digital campus, which relates to the field of digital technology and includes the following steps: collecting students' learning behavior data in real time to form a preliminary data set of students; analyzing the students' learning behavior in the preliminary data set to generate a personalized learning portrait for each student; determining the students' learning needs and resource demand characteristics based on the personalized learning portrait; pushing personalized learning paths and dynamically adjusting teaching resource allocation plans based on the students' learning needs and resource demand characteristics; collecting energy consumption data from each area, combining resource allocation results with the energy consumption data of each area, automatically optimizing energy consumption usage strategies, and obtaining an optimized learning environment. The resource scheduling strategy of the present invention, based on personalized needs, can not only improve students' learning efficiency but also ensure full utilization of resources.

Description

Technical Field

[0001] The present invention relates to the field of digital technology, and in particular to a comprehensive management system for a smart digital campus. Background Art

[0002] With the development of smart campuses and educational informatization, more and more universities and educational institutions have begun to introduce digital management systems to improve the utilization efficiency of teaching resources and optimize the learning environment. However, existing smart campus systems mostly focus on the static allocation and management of teaching resources, and lack the ability to dynamically respond to students' personalized learning needs. In the existing technology, although some systems can collect students' learning behavior data, most are limited to simple data statistical analysis, failing to deeply explore students' learning behavior patterns, and failing to conduct timely resource scheduling and learning path adjustments based on students' real-time performance and changes in the learning environment. In addition, the allocation of teaching resources is usually separated from the campus energy consumption management system, failing to achieve the coordinated optimization of resources and energy consumption, resulting in energy waste.

[0003] The shortcomings of existing technologies are mainly reflected in the following two aspects: First, the existing resource scheduling system lacks the ability to dynamically respond to students' personalized learning needs and cannot flexibly allocate resources based on students' real-time performance and changes in learning paths; second, the existing energy consumption management system is disconnected from the teaching resource scheduling system, and fails to optimize energy consumption based on real-time teaching resource usage, resulting in unnecessary energy waste. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a comprehensive management system for a smart digital campus to solve the problem of how to dynamically adjust the allocation of teaching resources and optimize energy consumption according to students' personalized learning needs and changes in the learning environment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] The present invention provides a comprehensive management system for a smart digital campus, which includes:

[0008] The data collection module is used to collect students’ learning behavior data in real time to form a preliminary data set of students;

[0009] The behavior analysis module is used to analyze the learning behavior of students in the preliminary data set and generate a personalized learning profile for each student;

[0010] The demand confirmation module determines students' learning needs and resource demand characteristics based on personalized learning portraits;

[0011] The resource push module is used to push personalized learning paths and dynamically adjust teaching resource allocation plans based on students' learning needs and resource requirements;

[0012] The optimization module is used to collect energy consumption data from each area, combine resource allocation results with energy consumption data from each area, and automatically optimize energy consumption strategies to obtain an optimized learning environment.

[0013] The adjustment module tracks students' learning progress and efficiency in real time through an optimized learning environment, and dynamically adjusts the learning path based on students' personalized learning profiles and current learning environment.

[0014] As a preferred solution of the integrated management system for smart digital campuses described in the present invention, wherein: the learning behavior data includes student classroom performance data, online learning data, achievement data and interaction data;

[0015] Clean and format the collected student behavior data;

[0016] The processed student behavior data is divided into multiple dimensions and indexed according to student ID, course ID and timestamp to form a preliminary data set of students.

[0017] As a preferred solution of the integrated management system for smart digital campus described in the present invention, the following steps are included in analyzing the learning behavior of students in the preliminary data set to form a student behavior feature vector:

[0018] Extracting concentration from students' facial expressions and body posture data as the main feature of their classroom performance;

[0019] Convert the online learning duration and frequency of students in the online learning data into continuous variables to obtain the learning duration and frequency features;

[0020] Convert the test scores and homework completion rates in the performance data into standard scores to form performance characteristics;

[0021] Generate quantitative features through the number of students participating in class and the quality of interaction in the interaction data;

[0022] All the extracted features are integrated to form the student behavior feature vector.

[0023] As a preferred solution of the integrated management system for smart digital campuses described in the present invention, generating a personalized learning profile of each student based on the student behavior feature vector includes the following steps:

[0024] Through the unsupervised K-means clustering algorithm, students are divided into different learning type groups according to their learning behavior feature vectors;

[0025] Use supervised learning with a random forest classifier, combined with students’ test scores and learning frequency labels, to predict each student’s learning ability and study habits;

[0026] Through K-means clustering and random forest classification results, a personalized learning portrait is generated for each student.

[0027] As a preferred solution of the integrated management system for smart digital campus described in the present invention, determining the learning needs of students based on personalized learning portraits includes the following steps:

[0028] Extract students’ learning habits and learning abilities from their personalized portraits;

[0029] Based on students’ learning habits and abilities, determine the intensity of students’ learning needs at the current stage;

[0030] Based on the intensity of students' learning needs, the urgency of students' learning and resource requirements are evaluated to obtain the characteristics of students' learning needs.

[0031] As a preferred solution of the integrated management system for smart digital campus described in the present invention, determining the learning resource demand characteristics of students based on personalized learning portraits includes the following steps:

[0032] Extract learning preferences from each student’s personalized profile;

[0033] Recommend learning resource types to students based on their learning preferences;

[0034] Recommend the best resource type based on students' preferences for different resource types and the availability of current resource libraries;

[0035] Determine the amount of resources students need based on their needs;

[0036] The optimal resource type and the required resource quantity are integrated to obtain the resource demand characteristics.

[0037] As a preferred solution of the integrated management system for smart digital campus described in the present invention, wherein: according to the learning demand characteristics and resource demand characteristics of students, pushing personalized learning paths and dynamically adjusting the teaching resource allocation plan includes the following steps:

[0038] Determine the subject areas that need to be prioritized based on the intensity of students’ learning needs;

[0039] Based on the student's resource type needs, select resources that meet this type of needs within the identified priority subject areas;

[0040] By calculating the similarity between student needs and the screened resource demand characteristics, the matching degree between student needs and the screened resource demand characteristics is determined;

[0041] Check the real-time availability of each resource from the matching results, exclude currently unavailable resources, and obtain the final available resources;

[0042] Generate personalized learning paths based on the final available resources, combined with the students' learning needs and resource demand characteristics;

[0043] Real-time detection of resource availability, rescheduling of final available resources, and adjustment of personalized learning paths are performed to obtain a dynamically adjusted teaching resource scheduling and allocation plan.

[0044] As a preferred solution of the integrated management system for smart digital campuses described in the present invention, wherein: the energy consumption data includes power consumption, equipment status and environmental parameters;

[0045] The power consumption refers to the real-time power consumption of classrooms and laboratories; the equipment status refers to whether the lighting system, air conditioning and experimental equipment are turned on; the environmental parameters refer to the temperature, humidity and light intensity data that affect students' learning comfort.

[0046] As a preferred solution of the integrated management system for smart digital campus described in the present invention, wherein: combining the resource allocation results and the energy consumption data of each area, automatically optimizing the energy consumption usage strategy, and obtaining the optimized learning environment includes the following steps:

[0047] Detect the environmental status and equipment usage of classrooms and laboratory areas based on energy consumption data;

[0048] Based on the results of teaching resource scheduling and allocation, determine whether the current area needs to maintain the current energy consumption status;

[0049] Automatically generate energy consumption optimization strategies based on detection and judgment results;

[0050] The energy consumption changes in each area are detected in real time. Based on the generated energy consumption optimization strategy, the power and switching status of the physical equipment are dynamically adjusted to obtain an optimized learning environment.

[0051] As a preferred solution of the integrated management system for smart digital campus described in the present invention, wherein: through the optimized learning environment, the learning progress and learning efficiency of students are tracked in real time, and the learning path is dynamically adjusted in combination with the personalized learning profile of the students and the current learning environment, including the following steps:

[0052] Track students' learning progress in real time based on their learning path, and obtain information on their progress in each learning step;

[0053] Evaluate students' learning efficiency through their progress in each learning step;

[0054] Evaluate students' learning outcomes based on their performance data in the learning path;

[0055] Dynamically adjust personalized learning paths based on students' learning performance and efficiency.

[0056] The beneficial effects of the present invention are: by collecting students' learning behavior data in real time, generating personalized learning portraits, and combining with the current learning environment, dynamically adjusting students' learning paths and teaching resource allocation plans; resource scheduling strategies based on personalized needs can not only improve students' learning efficiency, but also ensure full utilization of resources; combining energy consumption management with resource scheduling, utilizing energy consumption data from each area, automatically optimizing energy consumption usage strategies, and dynamically adjusting the power and switching status of physical equipment, thereby minimizing energy waste while ensuring the comfort of the learning environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 This is a flow chart of the comprehensive management system for smart digital campus in Example 1.

[0059] Figure 2 Schematic diagram of cluster distribution in Example 1. DETAILED DESCRIPTION

[0060] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0061] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0062] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0063] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a comprehensive management system for a smart digital campus, including a data collection module, a behavior analysis module, a demand confirmation module, a resource push module, an optimization module and an adjustment module.

[0064] Specifically, the data collection module is used to collect students' learning behavior data in real time to form a preliminary data set of students.

[0065] Learning behavior data includes student classroom performance data, online learning data, grade data, and interaction data; the collected student behavior data is cleaned and formatted; the processed student behavior data is divided into multiple dimensions and indexed according to student ID, course ID, and timestamp to form a preliminary data set for students.

[0066] Specifically, classroom performance data includes students' attention and participation in class. Facial expressions, eye tracking, body posture and other data collected by cameras are analyzed in combination with image recognition algorithms to determine students' concentration and participation. Online learning data includes the length of study, frequency of study, and course selection after students log in to the learning management system. The system records the timestamp of each student login, browsed learning resources, completed tasks, etc. Grade data includes students' test results and homework submission status. For subjective questions, the system uses an automatic scoring algorithm combined with teacher feedback to generate the final grade. Interaction data refers to the frequency and content of students' speeches collected through microphones, combined with natural language processing (NLP) algorithms to analyze the quality of students' language interaction.

[0067] The behavior analysis module is used to analyze the learning behavior of students in the preliminary data set and generate a personalized learning profile for each student.

[0068] Analyzing the learning behavior of students in the preliminary data set and forming the student behavior feature vector includes the following steps:

[0069] Concentration is extracted from the facial expression and body posture data in the classroom performance data as the main feature of students' classroom performance; the online learning time and frequency of students in the online learning data are converted into continuous variables to obtain learning time and frequency features; the test scores and homework completion rates in the performance data are converted into standard scores to form performance features; quantitative features are generated through the number of times students participate in class and the quality of interaction in the interaction data; all extracted features are integrated to form a student behavior feature vector.

[0070] Based on the student behavior feature vector, generating a personalized learning profile for each student includes the following steps:

[0071] Through the unsupervised K-means clustering algorithm, students are divided into different learning type groups according to their learning behavior feature vectors;

[0072] Use supervised learning with a random forest classifier, combined with students’ test scores and learning frequency labels, to predict each student’s learning ability and study habits;

[0073] Through K-means clustering and random forest classification results, a personalized learning portrait is generated for each student.

[0074] Specifically, the specific operations of the K-means clustering algorithm are:

[0075] Divide students into Each class represents a group of students with similar learning behaviors and habits. The feature vectors of each student are used as the initial cluster centers (centroids). These centroids represent the center points of each cluster. By minimizing the sum of the squares of the distances from each student to its cluster center, the students are divided into multiple different groups to form a learning profile for each student. The expression is:

[0076] ;

[0077] in, represents the sum of squared distances from all students to their cluster centers, represents the clustering index, represents the number of clusters, Indicates the The learning behavior feature vector of data points, represents the data point index, Indicates the The set of all data points in a cluster contains all data points belonging to the cluster (student behavior feature vectors). Indicates the The center point of a cluster.

[0078] Cluster 1 (learning progress): Students in this group have a higher learning time and frequency, their test scores gradually improve, and they behave actively in class.

[0079] Cluster 2 (Learning Delay): Students in this group study less, have lower test scores, and have lower class participation.

[0080] Cluster 3 (Self-driven learning): Students in this group study for an average amount of time, but study frequently, have high levels of participation, and achieve good grades.

[0081] Cluster 4 (passive learning): Students in this group spend a long time studying, but their learning outcomes are poor and their class participation is low.

[0082] Specifically, the operation of random forest classification is as follows:

[0083] Through random sampling and feature selection, multiple decision trees are constructed, each of which is generated based on a different subset of features and data. The student's behavioral feature vectors and clustering results are used as training data. Bootstrap sampling is used on the training data to generate multiple data sets. When each tree splits, the optimal feature is selected from a random subset of multiple feature data sets. For each decision tree, a classification model is generated using the training data, and each tree learns to classify learning ability or learning habits based on the features in the training data. The random forest integrates the results of multiple decision trees and ultimately obtains the prediction result through a voting mechanism, which is expressed as:

[0084] ;

[0085] in, Represents the final classification result, that is, the random forest model for the input feature vector (Students’ learning behavior characteristics) generate prediction results, Represents the majority voting mechanism, which selects the category with the most occurrences in the prediction results of all trees as the final classification result. Indicates the first decision tree for the input The classification results, Indicates that the second decision tree responds to the input The classification results, Indicates the A decision tree for input The classification results, Indicates the number of decision trees.

[0086] The demand confirmation module determines students' learning needs and resource demand characteristics based on personalized learning portraits.

[0087] Based on the personalized learning profile, determining students’ learning needs includes the following steps:

[0088] Extract students' learning habits and learning abilities from their personalized portraits; judge the intensity of students' learning needs at the current stage based on their learning habits and learning abilities; and evaluate the urgency and resource needs of students' learning based on the intensity of their learning needs to obtain the characteristics of their learning needs.

[0089] Based on the personalized learning portrait, determining the characteristics of students' learning resource needs includes the following steps:

[0090] Extract learning preferences from each student's personalized portrait; recommend learning resource types to students based on their learning preferences; recommend the optimal resource type based on students' preferences for different resource types and the availability of the current resource library; determine the number of resources required by students based on their demand characteristics; integrate the optimal resource type with the required resource quantity to obtain resource demand characteristics.

[0091] The resource push module is used to push personalized learning paths and dynamically adjust the teaching resource allocation plan based on students' learning needs and resource requirements. It includes the following steps:

[0092] Based on the intensity of students' learning needs, determine the subject areas that need to be matched first; based on the students' resource type needs, screen out resources that meet this type of needs within the determined priority subject areas; by calculating the similarity between students' needs and the screened resource demand characteristics, judge the degree of match between students' needs and the screened resource demand characteristics; check the real-time availability of each resource from the matching results, and exclude currently unavailable resources to obtain the final available resources; based on the final available resources, combine the students' learning demand intensity and resource demand characteristics to generate personalized learning paths; detect the available status of resources in real time, reschedule the final available resources, and adjust the personalized learning paths to obtain a dynamically adjusted teaching resource scheduling and allocation plan.

[0093] Specifically, the expression of similarity is:

[0094] ;

[0095] in, Indicates the The similarity between resources and student needs, similarity The larger the value, the higher the degree of matching between the resource and the student’s needs. represents the distance between student demand characteristics and resource demand characteristics, Represents the characteristics of student needs, Represents resource demand characteristics.

[0096] The expression of the distance between student demand characteristics and resource demand characteristics is:

[0097] ;

[0098] in, Indicates that students’ demand characteristics are The value of the dimension, Indicates the resource demand characteristics in The value of the dimension, represents the dimension of the feature vector, Represents the dimension index of the feature vector.

[0099] The expression generated by the personalized learning path is:

[0100] ;

[0101] in, represents a personalized learning path, represents the number of learning steps (i.e., the number of resources) contained in the learning path, represents the step index in the learning path, Indicates the Resources in each learning step, Indicates the The learning time or task time of each resource.

[0102] The optimization module is used to collect energy consumption data of each area, combine the resource allocation results and the energy consumption data of each area, automatically optimize the energy consumption usage strategy, and obtain an optimized learning environment.

[0103] Energy consumption data includes power consumption, equipment status and environmental parameters.

[0104] Power consumption refers to the real-time power consumption of classrooms and laboratories; equipment status refers to whether the lighting system, air conditioning and experimental equipment are turned on; environmental parameters refer to temperature, humidity and light intensity data that affect students' learning comfort.

[0105] Combining the resource allocation results and the energy consumption data of each area, the energy consumption strategy is automatically optimized. The optimized learning environment includes the following steps:

[0106] Based on energy consumption data, the environmental status and equipment usage of classroom and laboratory areas are detected; based on the results of teaching resource scheduling and allocation, it is determined whether the current area needs to maintain the existing energy consumption status; based on the detection and judgment results, an energy consumption optimization strategy is automatically generated; the energy consumption changes in each area are detected in real time, and based on the generated energy consumption optimization strategy, the power and switching status of physical equipment are dynamically adjusted to obtain an optimized learning environment.

[0107] The adjustment module tracks students’ learning progress and efficiency in real time through the optimized learning environment, and dynamically adjusts the learning path based on the students’ personalized learning profile and current learning environment. The steps include:

[0108] Track students' learning progress in real time based on their learning path, and obtain information on their progress in each learning step;

[0109] Evaluate students' learning efficiency through their progress in each learning step;

[0110] Evaluate students' learning outcomes based on their performance data in the learning path;

[0111] Dynamically adjust personalized learning paths based on students' learning performance and efficiency.

[0112] Specifically, the student's progress in each learning step includes the following data: whether the student completes the current learning step on time; the student's performance in the current learning step (such as test scores, homework accuracy, experiment success rate, etc.); and the time the student spends in the learning process.

[0113] Learning efficiency assessment means, for example, if a student completes 20 math problems within 1 hour with a high accuracy rate, then his or her learning efficiency is judged to be high.

[0114] Learning outcome assessment refers to determining whether students are performing well or having difficulty in a course based on data such as their scores on tests and the accuracy of their assignments.

[0115] Dynamic adjustment of personalized paths means that if a student performs well in basic mathematics courses (scores consistently above 90%), he or she will be recommended to study more difficult advanced mathematics courses; if a student makes slow progress in a course (for example, low test scores, high error rates in homework), additional learning resources or tutoring courses will be recommended to him or her.

[0116] In summary, the present invention generates personalized learning portraits by collecting students' learning behavior data in real time, and dynamically adjusts students' learning paths and teaching resource allocation plans based on the current learning environment; the resource scheduling strategy based on personalized needs can not only improve students' learning efficiency, but also ensure the full utilization of resources; it combines energy consumption management with resource scheduling, utilizes energy consumption data of each area, automatically optimizes energy consumption usage strategies, and dynamically adjusts the power and switching status of physical equipment, thereby minimizing energy waste while ensuring the comfort of the learning environment.

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions 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 preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A comprehensive management system for a smart digital campus, characterized in that: It includes data collection module, behavior analysis module, demand confirmation module, resource push module, optimization module and adjustment module; The data collection module is used to collect students' learning behavior data in real time to form a preliminary data set of students; The learning behavior data includes student classroom performance data, online learning data, performance data and interaction data; the collected student behavior data is cleaned and formatted; The processed student behavior data is divided into multiple dimensions and indexed according to student ID, course ID and timestamp to form a preliminary student data set; The classroom performance data includes students' attention and participation in class. Facial expressions, eye tracking, and body posture data collected by cameras are analyzed in combination with image recognition algorithms to determine students' attention and participation. The online learning data includes the length of time students study, learning frequency, and course selection after logging into the learning management system. The system records the timestamp of each student login, the learning resources browsed, and the tasks completed. The performance data includes students' test results and homework submission status. For subjective questions, the system uses an automatic scoring algorithm combined with teacher feedback to generate the final grade. The interaction data refers to the frequency and content of students' speeches collected through microphones, combined with natural language processing (NLP) algorithms to analyze the quality of students' language interaction. The behavior analysis module is used to analyze the learning behavior of students in the preliminary data set and generate a personalized learning profile for each student; Analyze the learning behavior of students in the preliminary data set to form student behavior feature vectors The method includes the following steps: extracting concentration from facial expressions and body posture data in students' classroom performance data as the main feature of students' classroom performance; converting students' online learning time and frequency in online learning data into continuous variables to obtain learning time and frequency features; converting test scores and homework completion rates in performance data into standard scores to form performance features; generating quantitative features through the number of students' class participation and the quality of interaction in interaction data; and integrating all extracted features to form a student behavior feature vector. Generating a personalized learning profile for each student based on their behavioral feature vectors involves the following steps: Using an unsupervised K-means clustering algorithm, students are divided into different learning type groups based on their learning behavior feature vectors; supervised learning using a random forest classifier, combined with their test scores and learning frequency labels, predicts each student's learning ability and learning habits; Generate a personalized learning profile for each student through K-means clustering and random forest classification results; Specifically, the K-means clustering algorithm operates as follows: students are divided into k categories, each category representing a group of students with similar learning behaviors and habits. The feature vector of each student is used as the initial cluster center, i.e., the centroid. These centroids represent the center point of each cluster. By minimizing the sum of the squares of the distances from each student to its cluster center, students are divided into multiple different groups to form a learning profile for each student. The expression is: Where J represents the sum of the squared distances from all students to their cluster centers, i represents the cluster index, k represents the number of clusters, and x represents the sum of the squared distances from all students to their cluster centers. j represents the learning behavior feature vector of the jth data point, j represents the data point index, C i Represents the set of all data points in the i-th cluster, including all data points belonging to the cluster, that is, the student behavior feature vector, μ i Represents the center point of the i-th cluster; Cluster 1, progressive learners: Students in this group have a higher study time and frequency, their test scores gradually improve, and they are active in class. Cluster 2, learning lag type: students in this group have fewer study hours, lower test scores, and lower class participation; Cluster 3, self-driven learners: students in this group study for an average amount of time, but study frequently, have high participation, and achieve good results; Cluster 4, passive learning: students in this group spend a long time studying, but their learning outcomes are poor and their class participation is low; Specifically, the operation of random forest classification is as follows: through random sampling and feature selection, multiple decision trees are constructed, and each tree is generated based on a different feature subset and data subset; the student's behavioral feature vector and clustering results are used as training data; bootstrap sampling is used on the training data to generate multiple data sets; when each tree splits, the optimal feature is selected from a random subset of multiple feature data sets; for each decision tree, a classification model is generated using the training data, and each tree learns to classify learning ability or learning habits based on the features in the training data; random forest integrates the results of multiple decision trees and finally obtains the prediction result through a voting mechanism, which is expressed as: H(X)=mode[h1(X),h2(X),...,h m (X)]; Among them, H(X) represents the final classification result, that is, the prediction result generated by the random forest model for the input feature vector X (students' learning behavior characteristics), mode represents the majority voting mechanism, and selects the category with the most occurrences in the prediction results of all trees as the final classification result, h1(X) represents the classification result of the first decision tree for input X, h2(X) represents the classification result of the second decision tree for input X, and h m (X) represents the classification result of the mth decision tree for input X, and m represents the number of decision trees; The demand confirmation module determines the student's learning needs and resource demand characteristics based on the personalized learning profile; Determining students' learning needs based on personalized learning profiles involves the following steps: extracting students' learning habits and learning abilities from each student's personalized profile; judging the intensity of students' learning needs at the current stage based on their learning habits and learning abilities; and assessing the urgency and resource requirements of students' learning based on the intensity of their learning needs, thereby obtaining the characteristics of their learning needs. Determining students' learning resource demand characteristics based on personalized learning profiles includes the following steps: extracting learning preferences from each student's personalized profile; recommending learning resource types based on the student's learning preferences; recommending the optimal resource type based on the student's preferences for different resource types and the availability of the current resource library; determining the number of resources required by the student based on the student's demand characteristics; and integrating the optimal resource type with the required resource quantity to obtain the resource demand characteristics. The resource push module is used to push personalized learning paths and dynamically adjust the teaching resource allocation plan based on the student's learning demand characteristics and resource demand characteristics, including the following steps: based on the student's learning demand intensity, determine the subject areas that need to be matched first; based on the student's resource type demand, screen out resources that meet the type of demand in the determined priority subject areas; by calculating the similarity between the student's demand and the screened resource demand characteristics, determine the degree of match between the student's demand and the screened resource demand characteristics; check the real-time availability of each resource from the matching results, and exclude currently unavailable resources to obtain the final available resources; based on the final available resources, combine the student's learning demand intensity and resource demand characteristics, and generate a personalized learning path; Real-time detection of resource availability, rescheduling of final available resources, and adjustment of personalized learning paths to obtain a dynamically adjusted teaching resource scheduling and allocation plan; Specifically, the expression of similarity is: Among them, S n It represents the similarity between the nth resource and the student's needs. The larger the similarity value, the higher the matching degree between the resource and the student's needs. stu , x res ) represents the distance between student demand characteristics and resource demand characteristics, x stu represents the student demand characteristics, x res Represents resource demand characteristics; The expression of the distance between student demand characteristics and resource demand characteristics is: Among them, x b Indicates the value of the student demand characteristic in the bth dimension, y b Indicates the value of the resource demand feature in the bth dimension, B represents the dimension of the feature vector, and b represents the dimension index of the feature vector; The expression generated by the personalized learning path is: Among them, P lpath represents the personalized learning path, N ste represents the number of learning steps contained in the learning path, that is, the number of resources, a represents the step index in the learning path, R a represents the resource in the a-th learning step, R a Indicates the learning time or task time of the ath resource; The optimization module is used to collect energy consumption data from each area, and automatically optimize the energy consumption strategy by combining the resource allocation results and the energy consumption data of each area to obtain an optimized learning environment. Energy consumption data includes power consumption, equipment status, and environmental parameters; Power consumption refers to the real-time power consumption of classrooms and laboratories; equipment status refers to whether the lighting system, air conditioning, and experimental equipment are turned on; environmental parameters refer to temperature, humidity, and light intensity data that affect students' learning comfort; Combining resource allocation results with energy consumption data for each area, the energy consumption strategy is automatically optimized to obtain an optimized learning environment. The following steps are involved: Based on energy consumption data, the environmental status and equipment usage of classroom and laboratory areas are detected; Based on the teaching resource scheduling and allocation results, it is determined whether the current area needs to maintain the existing energy consumption status; Based on the detection and judgment results, an energy consumption optimization strategy is automatically generated; Energy consumption changes in each area are detected in real time, and based on the generated energy consumption optimization strategy, the power and switching status of physical equipment are dynamically adjusted to obtain an optimized learning environment. The adjustment module uses the optimized learning environment to track students' learning progress and efficiency in real time, and dynamically adjusts the learning path based on the student's personalized learning profile and current learning environment. The module includes the following steps: tracking students' learning progress in real time based on each student's learning path, obtaining the student's progress in each learning step; evaluating students' learning efficiency based on their progress in each learning step; evaluating students' learning effects based on their performance data in the learning path; and dynamically adjusting personalized learning paths based on students' learning performance and efficiency. Specifically, the student's progress in each learning step includes the following data: whether the student completes the current learning step on time; how well the student performs in the current learning step; the time the student spends in the learning process; Learning efficiency assessment means that if a student completes a set number of math problems within a specified time and has a high accuracy rate, then their learning efficiency is considered high. Learning effectiveness assessment refers to determining whether students are performing well or having difficulties in a course based on their test scores and the accuracy of their assignments. Dynamically adjusting personalized pathways means that if a student performs well in a foundational math course, with scores consistently above 90%, they will be recommended to take a more difficult advanced math course. If a student makes slow progress in a course, with low test scores and a high error rate in assignments, additional learning resources or tutoring courses will be recommended. The above system generates personalized learning portraits by collecting students' learning behavior data in real time, and dynamically adjusts students' learning paths and teaching resource allocation plans based on the current learning environment; resource scheduling strategies based on personalized needs can not only improve students' learning efficiency, but also ensure full utilization of resources; energy consumption management is combined with resource scheduling, and energy consumption data from each area is used to automatically optimize energy consumption usage strategies and dynamically adjust the power and switching status of physical equipment, thereby minimizing energy waste while ensuring the comfort of the learning environment.

Images