Educational administration intelligent management method and system, and electronic equipment

Abstract

The invention relates to the technical field of educational administration intelligent management scheme design, in particular to an educational administration intelligent management method and system and electronic equipment. Comprising a data acquisition and preprocessing module, an intelligent analysis and decision-making module, a system integration and management module and the like. Data collection is comprehensive, preprocessing is fine, and multiple algorithm mining data values are covered. Intelligent analysis and decision relate to accurate recommendation of student courses, efficient scheduling of courses and classrooms, scientific monitoring of teaching quality and the like. The system can integrate an existing education system and is convenient to manage. According to the scheme, through a series of advanced technologies and innovative designs, many problems of traditional educational administration are effectively solved, and the intelligent level and the comprehensive quality of educational administration are comprehensively improved. In addition, the real-time performance and the accuracy are both considered, the scheme can be adjusted according to real-time data, and it is guaranteed that the management decision always fits the reality. According to the scheme, the educational administration management efficiency and quality are comprehensively improved, and great convenience is brought to educational management.

Description

Technical Field

[0001] The present invention relates to the technical field of design of intelligent teaching management solutions, and in particular to an intelligent teaching management method and system, and electronic equipment. Background Art

[0002] In the existing technology, there are some unresolved technical problems in the intelligent management solution of teaching affairs. For example: 1. Low data processing efficiency: With the continuous expansion of the scale of education, the amount of information about students, teachers and courses is huge. The traditional education management system has limitations in data storage, query and analysis, and cannot process massive data quickly and accurately, resulting in delayed management decisions.

[0003] 2. Unreasonable intelligent scheduling: In terms of course arrangement and classroom allocation, existing technologies can often only consider a single factor and cannot comprehensively consider various constraints, resulting in unreasonable scheduling results, affecting teaching order and student learning outcomes.

[0004] Therefore, the prior art needs to be further developed. Summary of the invention

[0005] The purpose of the present invention is to overcome the above technical deficiencies and provide an intelligent teaching management method and system, and electronic equipment to solve the problems existing in the prior art.

[0006] To achieve the above technical objectives, according to a first aspect of the present invention, the present invention provides an intelligent teaching management method, comprising: S1. Data collection and preprocessing: Through the data collection module, the school's existing system interface interacts and the sensor collects data, cleans the collected data, unifies the format and standardizes it, and performs feature extraction and feature engineering; S2. Student course recommendation: Based on students’ historical learning data and course information, a collaborative filtering algorithm and a content-based recommendation algorithm are combined to build a student course recommendation model, set a recommendation score threshold, and recommend courses with scores higher than the threshold to students; S3, Course Arrangement and Classroom Scheduling: Construct a course arrangement and classroom scheduling model, use a combination of genetic algorithm and simulated annealing algorithm to solve it, set the classroom capacity utilization threshold, and adjust the classroom allocation plan according to the threshold; S4. Teaching evaluation and quality monitoring: Design a teaching evaluation index system, collect multi-party evaluation data, use hierarchical analysis method and fuzzy comprehensive evaluation method to calculate the comprehensive score of teachers' teaching quality, set the qualified threshold for teaching quality evaluation, and provide improvement suggestions and training plans for teachers whose scores are below the threshold.

[0007] Specifically, in the data collection and preprocessing steps, the collected data includes student information, teacher information and course information. The student information includes student number, name, major, and grade. The teacher information includes work number, name, title, and teaching subjects. The course information includes course number, course name, credits, class time, and class location.

[0008] Specifically, in the student course recommendation step, the collaborative filtering algorithm calculates the similarity between students and makes recommendations based on the courses selected by similar students, and the content-based recommendation algorithm makes matching recommendations by analyzing the course content and the students' majors and interests.

[0009] Specifically, the specific method for the student to recommend a course includes: S21. Course content feature extraction: Preprocess the text information of the syllabus, textbook content, and course introduction of each course, including word segmentation, stop word removal, and stem extraction operations; convert the preprocessed text into a vector representation using a word vector model, and calculate the vector value of each word. The word vector model includes Word2Vec or GloVe; obtain the feature vector of the course content by calculating the average value and weighted average value of the word vectors in the text, which can reflect the core knowledge and theme of the course; S22. Student interest modeling: Collect students’ behavior data during the learning process, including course browsing records, course selection history, study time, and homework completion status; Based on students’ behavior data, use association rule mining or cluster analysis methods to discover students’ interest preferences and potential needs; Express students’ interest preferences in vector form, which has the same dimension as the course content feature vector; S23, matching recommendation calculation: Calculate the similarity between the student interest vector and the course content feature vector, using the cosine similarity or Euclidean distance similarity measurement method; sort the courses according to the similarity score, and select the courses with similarity higher than the recommended score threshold as recommended courses.

[0010] Specifically, in the course arrangement and classroom scheduling step, the course time requirement, classroom capacity, multimedia equipment requirements, and teacher and student free time are input into the course arrangement and classroom scheduling model as constraints.

[0011] Specifically, in the course arrangement and classroom scheduling steps, the neighborhood search process of the simulated annealing operation includes: Validity check and correction of neighborhood solutions: After generating a neighborhood solution, check the validity of the neighborhood solution to ensure that the new course arrangement and classroom allocation plan meet all constraints, including that the course time does not conflict and the classroom capacity meets the student number requirements. For neighborhood solutions that do not meet the constraints, take corresponding correction measures, including: if there is a course time conflict, reselect a suitable classroom to eliminate the conflict; if the classroom capacity is insufficient, replace it with a classroom that can accommodate the corresponding number of students.

[0012] Specifically, in the course arrangement and classroom scheduling steps, the crossover operation of the genetic algorithm adopts a multi-point crossover method, which specifically includes the following steps: Select crossover points: for the two selected parent chromosomes, randomly select k crossover points, where the value range of k is [2, chromosome length / 4], and the chromosome length represents the number of courses; Gene exchange: exchange genes of two parent chromosomes at the selected k crossover points to generate two new daughter chromosomes; Conflict handling: Check whether there is a conflict between courses and classrooms in the newly generated offspring chromosomes. If there is a conflict, a greedy algorithm is used to adjust it, that is, giving priority to meeting the course schedule and classroom capacity constraints, and exchanging or adjusting the classroom allocation of some courses until the conflict is eliminated.

[0013] Specifically, in the teaching evaluation and quality monitoring steps, the teaching evaluation index system includes teaching content, teaching methods, and teacher teaching attitude.

[0014] According to a second aspect of the present invention, there is provided an intelligent education management system, comprising: Data acquisition and preprocessing module: used to interact with the school's existing system interface and collect data from sensors through the data acquisition module, clean the collected data, unify the format and standardize it, and perform feature extraction and feature engineering; Student course recommendation module: Based on students' historical learning data and course information, a student course recommendation model is constructed by combining collaborative filtering algorithm and content-based recommendation algorithm, and a recommendation score threshold is set to recommend courses with scores higher than the threshold to students; Course arrangement and classroom scheduling module: build a course arrangement and classroom scheduling model, comprehensively consider multiple factors, use a combination of genetic algorithm and simulated annealing algorithm to solve, set the classroom capacity utilization threshold, and adjust the classroom allocation plan according to the threshold; Teaching evaluation and quality monitoring module: Design a teaching evaluation index system, collect multi-party evaluation data, use hierarchical analysis method and fuzzy comprehensive evaluation method to calculate the comprehensive score of teachers' teaching quality, set the qualified threshold for teaching quality evaluation, and provide improvement suggestions and training plans for teachers whose scores are below the threshold.

[0015] According to a third aspect of the present invention, there is provided an electronic device, comprising: a memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the above-mentioned intelligent teaching management method is implemented.

[0016] Beneficial effects: The intelligent teaching management solution of the present invention has many significant beneficial effects. First, data processing is efficient. Through advanced data preprocessing and mining technology, it can quickly analyze massive teaching data and provide timely support for management decisions. Secondly, intelligent scheduling is more reasonable. The optimization algorithm that comprehensively considers multiple factors ensures that courses and classroom arrangements are scientific and efficient, greatly improving the utilization rate of teaching resources. In addition, real-time and accuracy are given equal importance. The solution can be adjusted according to real-time data to ensure that management decisions are always in line with reality. This solution comprehensively improves the efficiency and quality of teaching management and brings great convenience to education management. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flow chart of the teaching affairs intelligent management method provided in a specific embodiment of the present invention; Figure 2 It is a schematic diagram of the system composition of the intelligent teaching management system provided in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings of the present invention. Based on the embodiments in this application, other similar embodiments obtained by ordinary technicians in this field without making creative work should all fall within the scope of protection of this application. In addition, the directional words mentioned in the following embodiments, such as "up", "down", "left", "right", etc., are only reference to the directions of the accompanying drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0019] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments.

[0020] See also Figure 1 The present invention provides an intelligent teaching management method, comprising: S1. Data collection and preprocessing: The data collection module interacts with the school's existing system interface and collects data through sensors, cleans the collected data, unifies the format and standardizes it, and performs feature extraction and feature engineering.

[0021] Specifically, in the data collection and preprocessing steps, the collected data includes student information, teacher information and course information. The student information includes student number, name, major, and grade. The teacher information includes work number, name, title, and teaching subjects. The course information includes course number, course name, credits, class time, and class location.

[0022] 1.1 Data Collection In a preferred embodiment, the specific process of data collection and preprocessing includes: designing a data collection module, interacting with the school's existing teaching management system, teaching resource management system and other interfaces to collect various types of teaching data, including but not limited to the following information: Student information: student ID, name, major, grade, enrollment date, grades of courses taken, etc.

[0023] Teacher information: work number, name, professional title, department, subjects taught, teaching evaluation scores, etc.

[0024] Course information: course number, course name, credits, class time (specific week and class), class location (classroom number), course nature (compulsory / elective), prerequisite course requirements, etc.

[0025] Teaching plan arrangement: course opening plan, teaching weeks, examination week arrangement, etc. for each major each semester. At the same time, sensors, cameras and other equipment are used to collect real-time data such as classroom usage (whether it is idle, currently occupied courses, etc.) and student attendance (sign-in time, attendance times, etc.) to obtain more comprehensive teaching information.

[0026] 1.2 Data Preprocessing 1.2.1 Data Cleaning Check the completeness and accuracy of data through data validation rules, and remove duplicate, erroneous or incomplete data records. For example: The ID card number in the student information will be verified according to the coding rules of the ID card number. Records that do not meet the standards will be considered as erroneous data and deleted.

[0027] Check whether the class time in the course information is reasonable, such as whether there is a time conflict or it exceeds the normal teaching time range, and correct or delete unreasonable data.

[0028] 1.2.2 Format unification and standardization The data from different sources are formatted and standardized for subsequent data processing and analysis. The specific conversion rules are as follows: Unify the students' major names into the standard names stipulated by the school, establish a major name mapping table, and replace major names that do not conform to the standard names with standard names.

[0029] Unify the time data into a specific time format (such as yyyy-MM-ddHH:mm:ss) and implement data format conversion by writing a time format conversion function.

[0030] 1.2.3 Feature extraction and feature engineering Extract valuable features from raw data and perform feature engineering to prepare for subsequent analysis and modeling. Specific operations include: Student feature extraction: Calculate students’ performance characteristics, such as average score (the average of all course scores), highest score, lowest score, standard deviation of scores, etc.; analyze students’ learning trends, such as changes in GPA per semester, etc.

[0031] Course feature extraction: Analyze the difficulty characteristics of the course, which can be comprehensively evaluated through students' average scores, score distribution, etc.; extract the relevance characteristics of the course, such as the prerequisite relationship between the course and other courses, the overlap of knowledge points, etc.

[0032] S2. Student course recommendation: Based on students’ historical learning data and course information, a student course recommendation model is constructed by combining collaborative filtering algorithm and content-based recommendation algorithm. A recommendation score threshold is set and courses with scores higher than the threshold are recommended to students.

[0033] Specifically, in the student course recommendation step, the collaborative filtering algorithm calculates the similarity between students and makes recommendations based on the courses selected by similar students, and the content-based recommendation algorithm makes matching recommendations by analyzing the course content and the students' majors and interests.

[0034] Specifically, the specific method for the student to recommend a course includes: S21. Course content feature extraction: Preprocess the text information of the syllabus, textbook content, and course introduction of each course, including word segmentation, stop word removal, and stem extraction operations; convert the preprocessed text into a vector representation using a word vector model, and calculate the vector value of each word. The word vector model includes Word2Vec or GloVe; obtain the feature vector of the course content by calculating the average value and weighted average value of the word vectors in the text, which can reflect the core knowledge and theme of the course; S22. Student interest modeling: Collect students' behavioral data during the learning process, including course browsing records, course selection history, learning time, and homework completion status; based on students' behavioral data, use association rule mining or cluster analysis methods to discover students' interest preferences and potential needs; express students' interest preferences in vector form, which has the same dimension as the course content feature vector.

[0035] Specifically, the specific method of student interest modeling includes: The basis of student interest modeling is the behavioral data of students in the learning process, which specifically includes the following: ①. Data collection Course browsing records: the number of times, duration, frequency, etc. that students browse courses in the academic affairs system.

[0036] Course Selection History: A list of courses a student has taken, including both elective and required courses.

[0037] Study time: the actual study time of students in each course (which can be recorded through the online learning platform).

[0038] Homework completion status: number of times students submit homework, homework grades, homework completion time, etc.

[0039] Examination results: students' examination results or regular grades in each course.

[0040] Classroom participation: number of student sign-ins in class, frequency of classroom interaction, etc. (can be collected through smart classroom equipment).

[0041] ②. Data preprocessing Clean and standardize the collected student behavior data to ensure the integrity and consistency of the data: Data cleaning: Remove invalid data: such as abnormal data such as negative learning time and 0 completed assignments.

[0042] Fill missing values: For missing study time or homework completion status, use mean filling or interpolation method to supplement.

[0043] Deduplication: Delete duplicate course browsing records.

[0044] Unified format: Standardize time data into a standard time format (such as yyyy-MM-dd HH:mm:ss).

[0045] Standardize study time into minutes.

[0046] Feature extraction: Calculate students' course preference characteristics, such as: Course type preference: the ratio of required courses to elective courses in the courses taken by students.

[0047] Course difficulty preference: Analyze students’ preference for high-difficulty courses or low-difficulty courses based on their grade distribution.

[0048] Course time preference: the time period distribution of students’ course selection (e.g. morning, afternoon, evening).

[0049] Course field preference: Statistics on students’ interest areas based on the field of the course (such as liberal arts, science, and engineering).

[0050] ③. Feature Engineering Convert student behavior data into feature vectors that can be used for modeling: Feature vector construction: Each student's interest feature vector is a multidimensional vector, and the dimension is determined by the specific features. For example: Feature 1: Number of course views (normalized to the range [0,1]).

[0051] Feature 2: Learning duration (normalized to the range [0,1]).

[0052] Feature 3: Assignment completion rate (number of completions / total number of courses).

[0053] Feature 4: Average test score (normalized to the range [0,1]).

[0054] Feature 5: Preference for course type (such as the weight distribution of liberal arts, science, and engineering).

[0055] Assume that the feature vector is: Vs=[v1,v2,v3,v4,v5], where Vs represents the interest feature vector of student s.

[0056] Feature weight assignment: Assign weights based on the importance of the features. For example: The weight of study time is 0.4, the weight of test score is 0.3, the weight of course browsing times is 0.2, and the weight of homework completion rate is 0.1.

[0057] The weighted eigenvector is: Vs′=w1*v1+w2*v2+w3*v3+w4*v4+w5*v5.

[0058] ④. Interest Modeling Method Use association rule mining or cluster analysis methods to discover students' interest preferences and potential needs.

[0059] (4.1) Association rule mining Goal: To discover correlations between student behaviors, such as “students who like to take liberal arts courses also tend to like to take arts courses.”

[0060] Specific methods include: Use Apriori algorithm or FP-Growth algorithm to mine frequent itemsets.

[0061] Define support and confidence thresholds to filter out meaningful rules.

[0062] Example rule: "If a student chooses 'Introduction to Literature', there is an 80% chance that he or she will also choose 'Ancient Literature'."

[0063] Output: Generate a set of student interest association rules.

[0064] (4.2) Cluster analysis Goal: To divide students into different interest groups for easy follow-up recommendations.

[0065] method: Using K-Means clustering algorithm, students are divided into k interest groups.

[0066] The feature vector Vs′ is taken as input, and the clustering result is the label of the interest group to which the student belongs.

[0067] Example clustering results: Group 1: Students who prefer liberal arts courses.

[0068] Group 2: Students who prefer science courses.

[0069] Group 3: Students interested in multi-disciplinary courses.

[0070] Output: interest group label for each student.

[0071] ⑤. Interest vector representation The student's interest preference is expressed as a vector with the same dimension as the course content feature vector: Assuming that the course content feature vector is Vc=[c1,c2,c3,…,cn], the dimension of the student interest vector Vs′ is consistent with Vc after normalization and standardization.

[0072] Example: The course content feature vector Vc=[0.8,0.6,0.2] (indicating the course’s difficulty, relevance, time arrangement, and other characteristics).

[0073] Student interest vector Vs′=[0.7,0.5,0.3] (indicating students’ preferences for difficulty, relevance, and time arrangement).

[0074] ⑥. Data setting example The following is a specific data setting example: Student Behavior Data: Student A’s course browsing records: browsed “Introduction to Literature” 10 times, browsed “Ancient Literature” 8 times.

[0075] Student A's study time: The total study time is 20 hours, including 12 hours of "Introduction to Literature" and 6 hours of "Ancient Literature".

[0076] Student A's homework completion status: completion rate 90%, average test score 85 points.

[0077] Eigenvectors: Student A’s interest feature vector Vs′=[0.8, 0.7, 0.9, 0.85] (assuming it has been normalized and standardized).

[0078] Course content feature vector: The characteristic vector of the course "Ancient Literature" is Vc=[0.7,0.6,0.5].

[0079] ⑦. Similarity calculation The cosine similarity or Euclidean distance is used to calculate the similarity between the student interest vector and the course content feature vector: Cosine Similarity: .

[0080] Euclidean distance: .

[0081] Example calculation: Assume that student A’s interest vector Vs′=[0.8,0.7,0.9] and course feature vector Vc=[0.7,0.6,0.5].

[0082] Cosine Similarity: ⑧. Recommended threshold setting Set the recommendation score threshold T, for example, T=0.8.

[0083] If sim(Vs′,Vc)>T, the course is recommended to the student.

[0084] Through the above steps, the process of student interest modeling is refined into specific links such as data collection, preprocessing, feature engineering, interest modeling, vector representation, similarity calculation and recommendation threshold setting. Each link clarifies the source of data, processing method and calculation formula to ensure the scientificity and operability of the modeling process.

[0085] S23, matching recommendation calculation: Calculate the similarity between the student interest vector and the course content feature vector, using the cosine similarity or Euclidean distance similarity measurement method; sort the courses according to the similarity score, and select the courses with similarity higher than the recommended score threshold as recommended courses.

[0086] Furthermore, the specific method of the student recommending a course includes: Based on students’ historical learning data (such as grades of completed courses, course preferences) and relevant information of courses (course difficulty, course relevance, etc.), a student course recommendation model is constructed. Specifically, a collaborative filtering algorithm and a content-based recommendation algorithm are combined. The steps are as follows: 2.1 Student Course Recommendations 2.1.1 Algorithm steps Data preparation: Organize students’ historical learning data and course information into a data format suitable for algorithm processing, such as a matrix, where rows represent students, columns represent courses, and matrix elements are students’ course grades or other relevant features.

[0087] Collaborative filtering stage: Calculate student similarity: Use the cosine similarity algorithm to calculate the similarity between students. For any two students i and j, the similarity calculation formula is: Among them, sim(i,j) represents the similarity, x ik represents the score of student i in the kth course, and n is the number of courses that the two students take together.

[0088] Find neighbor students: According to the set number of neighbors N (for example, N=20, that is, select the 20 students most similar to the target student), find the N neighbor students with the highest similarity to the target student.

[0089] Predicted rating: For the courses that the target student has not chosen, the predicted rating is calculated based on the ratings and similarity of the neighboring students for the course. The calculation formula is: Among them, P ij represents the predicted score of target student i for course j, R kj represents the actual rating of neighbor student k for course j.

[0090] Content-based recommendation stage: Construct a course feature vector: Represent the relevant features of the course (such as course difficulty, course relevance, etc.) in vector form, with each dimension corresponding to a feature value.

[0091] Calculate course similarity: Use the Euclidean distance algorithm to calculate the similarity between courses. For any two courses m and n, the Euclidean distance calculation formula is: Among them, X lm represents the value of course m on the lth feature, and m is the dimension of the feature vector of course m; X ln Indicates the course n The value of the lth feature, n is the dimension of the feature vector for course n.

[0092] Recommended course selection: Based on the similarity between the target students’ courses and the candidate courses, select courses with higher similarity as recommended courses.

[0093] Comprehensive recommendation: weighted combination of the recommended courses obtained by the collaborative filtering algorithm and the content-based recommendation algorithm to obtain the final recommended course list. Let the score of the course recommended by collaborative filtering be Scf, the score of the course recommended by content be Scb, and the comprehensive score be S=w1×Scf+w2×Scb, where w1 and w2 are the weights of the two algorithms, and w1+w2=1. It can be understood that w1 and w2 can be specifically set according to the actual needs of the user of the present invention, as long as the sum of w1 and w2 is 1.

[0094] 2.1.2 Threshold Setting Set a recommendation score threshold, and only courses with a recommendation score higher than the threshold will be recommended to students. For example, when the recommendation score is greater than 8 points (out of 10 points), the course will be recommended to students.

[0095] S3. Course scheduling and classroom scheduling: Construct a course scheduling and classroom scheduling model, use a combination of genetic algorithm and simulated annealing algorithm to solve it, set the classroom capacity utilization threshold, and adjust the classroom allocation plan according to the threshold.

[0096] Specifically, in the course arrangement and classroom scheduling step, the course time requirement, classroom capacity, multimedia equipment requirements, and teacher and student free time are input into the course arrangement and classroom scheduling model as constraints.

[0097] Specifically, in the course arrangement and classroom scheduling steps, the neighborhood search process of the simulated annealing operation includes: Course classroom exchange strategy: Randomly select two different courses and exchange their currently assigned classrooms to form a new neighborhood solution. For example, if course A is currently scheduled in classroom 1 and course B is currently scheduled in classroom 2, then in the newly generated neighborhood solution, course A is scheduled in classroom 2 and course B is scheduled in classroom 1.

[0098] Course classroom reallocation strategy: randomly select a course, randomly select a classroom that is not occupied by other courses and meets the course classroom requirements from all available classrooms, reallocate the course to the new classroom, and keep the arrangement of other courses unchanged, so as to obtain a new neighborhood solution.

[0099] Local adjustment strategy: select a continuous course schedule (such as multiple courses on a certain day or courses within a certain time period) and make local adjustments to the classroom allocation of these courses. You can randomly select an adjustment method, such as exchanging the classroom of one course with the classroom of another course, or arranging one course to a new suitable classroom, while keeping the classroom arrangements of other courses unchanged, thus forming a new neighborhood solution.

[0100] Validity check and correction of neighborhood solutions: After generating a neighborhood solution, check the validity of the neighborhood solution to ensure that the new course arrangement and classroom allocation plan meet all constraints, including that the course time does not conflict and the classroom capacity meets the student number requirements. For neighborhood solutions that do not meet the constraints, take corresponding correction measures, including: if there is a course time conflict, reselect a suitable classroom to eliminate the conflict; if the classroom capacity is insufficient, replace it with a classroom that can accommodate the corresponding number of students.

[0101] Neighborhood solution screening and retention strategy: After obtaining multiple valid neighborhood solutions, select some neighborhood solutions that are more likely to guide the algorithm to converge to a better solution according to certain screening rules for further evaluation and processing. The screening rules can be based on the quality and diversity of neighborhood solutions. For example, give priority to neighborhood solutions that have a large difference in fitness value from the current solution but are still within an acceptable range. At the same time, it is also necessary to ensure that the selected neighborhood solutions have a certain diversity to avoid the algorithm from falling into a local optimal solution too early. For the neighborhood solutions retained after screening, enter the subsequent acceptance judgment step.

[0102] Specifically, in the course arrangement and classroom scheduling steps, the crossover operation of the genetic algorithm adopts a multi-point crossover method, which specifically includes the following steps: Select crossover points: for the two selected parent chromosomes, randomly select k crossover points, where the value range of k is [2, chromosome length / 4], and the chromosome length represents the number of courses; Gene exchange: exchange genes of two parent chromosomes at the selected k crossover points to generate two new daughter chromosomes; Conflict handling: Check whether there is a conflict between courses and classrooms in the newly generated offspring chromosomes. If there is a conflict, a greedy algorithm is used to adjust it, that is, giving priority to meeting the course schedule and classroom capacity constraints, and exchanging or adjusting the classroom allocation of some courses until the conflict is eliminated.

[0103] Furthermore, the specific methods of course arrangement and teacher scheduling include: The course arrangement and classroom scheduling model is constructed, and various factors are comprehensively considered. The genetic algorithm and simulated annealing algorithm are combined to solve the problem. The steps are as follows: 2.2 Course Arrangement and Classroom Scheduling 2.2.1 Algorithm steps Encoding: Encode the course schedule and classroom allocation scheme into chromosome form, and the genes on the chromosome represent the correspondence between courses and classrooms. For example, a chromosome can be represented as a vector with a length equal to the number of courses, and each element corresponds to a classroom number.

[0104] Initialize the population: Randomly generate an initial chromosome population, and set the population size to M (for example, M=100, that is, randomly generate 100 initial course schedules and classroom allocation plans).

[0105] Fitness evaluation: Define a fitness function to evaluate the quality of chromosomes. The factors considered by the fitness function include course time conflicts, whether the classroom capacity meets the requirements, and the efficiency of teaching resource utilization. For example, the fitness function can be expressed as: F=1 / (1+C+R) Among them, C represents the number of course time conflicts, and R represents the idle rate of classroom resources (the ratio of the number of idle classrooms to the total number of classrooms).

[0106] Genetic Operations: Selection: Roulette wheel selection algorithm is used to select parent chromosomes according to the fitness value of the chromosome. The higher the fitness value of the chromosome, the greater the probability of being selected.

[0107] Crossover: A crossover operation is used to exchange genes between the selected parent chromosomes to generate new daughter chromosomes. For example, a single-point crossover operation randomly selects a crossover point and exchanges genes between the two parent chromosomes at the crossover point to obtain two daughter chromosomes.

[0108] Mutation: Use mutation operations to mutate some genes in the offspring chromosomes to increase the diversity of the population. For example, randomly change the value of a gene with a certain mutation probability (such as 0.05) (that is, change the correspondence between courses and classrooms).

[0109] Simulated annealing operation: Heating process: Set the initial temperature T0 (for example, T0=100), and gradually reduce the temperature according to a certain temperature reduction coefficient (such as 0.95).

[0110] Neighborhood search: At the current temperature, a neighborhood search is performed on each chromosome in the population to generate new individuals by randomly changing one or more genes on the chromosome.

[0111] Acceptance criteria: For the new individual found, calculate its fitness value fnew and the fitness value fold of the current individual (parent individual), and decide whether to accept the new individual according to the Metropolis criterion. When fnew>fold, the new individual is accepted; when fnew≤fold, the new individual is accepted with probability e−Δf / T, where Δf=fold−fnew, and T is the current temperature.

[0112] Termination condition: When the set maximum number of iterations (such as 1000) is reached or the temperature drops below a certain threshold (such as 0.01), the algorithm is terminated to obtain the current optimal course schedule and classroom allocation plan.

[0113] 2.2.2 Threshold Setting Set classroom capacity utilization thresholds to ensure that classroom resources allocated to courses are fully utilized. For example, when classroom capacity utilization is less than 70%, readjust the classroom allocation plan.

[0114] S4. Teaching evaluation and quality monitoring: Design a teaching evaluation index system, collect multi-party evaluation data, use hierarchical analysis method and fuzzy comprehensive evaluation method to calculate the comprehensive score of teachers' teaching quality, set the qualified threshold for teaching quality evaluation, and provide improvement suggestions and training plans for teachers whose scores are below the threshold.

[0115] Specifically, in the teaching evaluation and quality monitoring steps, the teaching evaluation index system includes teaching content, teaching methods, and teacher teaching attitude.

[0116] Specifically, the specific steps of the teaching evaluation and quality control include: 2.3 Teaching Evaluation and Quality Monitoring 2.3.1 Algorithm steps Design a teaching evaluation index system, evaluate the teaching quality of teachers from multiple dimensions such as teaching content, teaching methods, and teachers' teaching attitudes, collect multi-party evaluation data, and use hierarchical analysis method and fuzzy comprehensive evaluation method to calculate the comprehensive score of teachers' teaching quality. The steps are as follows: Determine the weight of evaluation indicators: Use the analytic hierarchy process (AHP) to determine the weight of the teaching evaluation indicator system. First, build a hierarchical model to decompose the overall teaching evaluation goal into multiple sub-goals (such as teaching content, teaching methods, teaching attitude, etc.), and then further decompose each sub-goal into specific evaluation indicators. Then, obtain the relative importance of the evaluation indicators through expert scoring or questionnaire survey methods, build a judgment matrix, and calculate the maximum eigenvalue and corresponding eigenvector of the judgment matrix to obtain the weight vector W of each evaluation indicator.

[0117] Data collection and quantification: Collect evaluation data from students, peers and supervisors, and quantify the evaluation data. For example, use the fuzzy scoring method to divide the evaluation levels into excellent, good, medium, pass and fail, corresponding to the numerical intervals [90,100], [80,90), [70,80), [60,70) and [0,60) respectively.

[0118] Fuzzy comprehensive evaluation: The fuzzy comprehensive evaluation method is used to calculate the comprehensive score of teachers' teaching quality. According to the quantified evaluation data, the fuzzy evaluation matrix R is constructed, and the fuzzy evaluation results of teachers' teaching quality are obtained by calculating B=W×R (where B is the fuzzy comprehensive evaluation result vector). Finally, the fuzzy evaluation results are defuzzified to obtain the comprehensive score of teachers' teaching quality.

[0119] Furthermore, the method comprises: In the teaching evaluation and quality control step, when the analytic hierarchy process is used to determine the weight of the teaching evaluation index system, the following steps are specifically included: Construct a judgment matrix: collect the relative importance data between evaluation indicators through expert scoring method or questionnaire survey method. For each pair of evaluation indicators, determine the value of its relative importance according to the 1-9 scale method, and construct a judgment matrix A=(aij), where aij represents the importance scale value of indicator i relative to indicator j, and satisfies aij>0, aji=1 / aij, aii=1; Calculate the eigenvector and the maximum eigenvalue: Use the power method or the sum-product method to calculate the maximum eigenvalue λmax of the judgment matrix A and its corresponding eigenvector W, normalize the eigenvector W so that the sum of its elements is 1, and obtain the relative weight vector of each evaluation index; Consistency test: Calculate the consistency index CI = (λmax-n) / (n-1), where n is the order of the judgment matrix; find the average random consistency index RI (its value depends on the order n of the judgment matrix); calculate the consistency ratio CR = CI / RI. When CR < 0.1, it is considered that the judgment matrix has satisfactory consistency and the weight vector W is valid; otherwise, the judgment matrix needs to be readjusted until the consistency requirements are met; Total ranking of levels: Determine the combined weight of each evaluation indicator for the overall teaching evaluation goal. If the teaching evaluation indicator system is divided into m levels, the k-1th level has n k−1 elements, and the kth level has n k elements, and the weight of the k-th layer element for the j-th element in the k-1th layer is W jk , the weight of the jth element in the k−1th layer for the total target is , then the calculation formula for the combined weight of the i-th element in the k-th layer for the total target is as follows: ; The combined weights of each evaluation index for the overall teaching evaluation goal are calculated in a bottom-up manner to obtain the final weight distribution result.

[0120] 2.3.2 Threshold Setting Set a qualified threshold for teaching quality evaluation, and provide targeted improvement suggestions and training plans for teachers whose teaching evaluation scores are below the threshold. For example, when a teacher's comprehensive teaching quality score is below 60 points, notify the teacher to participate in a teaching training course.

[0121] Specifically, the present invention also provides a solution for intelligent reminder and notification, which specifically includes the following steps: Based on the school's teaching schedule and student information, notifications and reminders are sent to students and teachers in a variety of ways. The steps are as follows: Notification information generation: Generate corresponding notification and reminder information based on the school's teaching schedule (such as course change notifications, exam arrangement notifications, paper submission deadline reminders, etc.) and relevant information of students or teachers (such as class schedules, paper start time, etc.).

[0122] Information push: Notifications and reminders are sent to corresponding students and teachers according to the receiving method selected by the user (such as SMS, email or APP push).

[0123] It should be noted here that the intelligent teaching management solution of the present invention has many significant beneficial effects. First, data processing is efficient. Through advanced data preprocessing and mining technology, it can quickly analyze massive teaching data and provide timely support for management decisions. Secondly, intelligent scheduling is more reasonable. The optimization algorithm that comprehensively considers multiple factors ensures that courses and classroom arrangements are scientific and efficient, greatly improving the utilization rate of teaching resources. In addition, real-time and accuracy are given equal importance, and the solution can be adjusted according to real-time data to ensure that management decisions are always in line with reality. This solution comprehensively improves the efficiency and quality of teaching management, bringing great convenience to education management.

[0124] See also Figure 2 The present invention provides another embodiment, which provides an intelligent management system for teaching affairs, and the intelligent management system for teaching affairs includes: Data collection and preprocessing module 100: used to interact with the school's existing system interface and collect data through sensors through the data collection module, clean the collected data, unify the format and standardize the data, and perform feature extraction and feature engineering; Student course recommendation module 200: Based on the student's historical learning data and course information, a student course recommendation model is constructed by combining collaborative filtering algorithm and content-based recommendation algorithm, and a recommendation score threshold is set to recommend courses with scores higher than the threshold to students; Course arrangement and classroom scheduling module 300: construct a course arrangement and classroom scheduling model, comprehensively consider multiple factors, use a combination of genetic algorithm and simulated annealing algorithm to solve, set a classroom capacity utilization threshold, and adjust the classroom allocation plan according to the threshold; Teaching evaluation and quality monitoring module 400: Design a teaching evaluation index system, collect multi-party evaluation data, use hierarchical analysis method and fuzzy comprehensive evaluation method to calculate the comprehensive score of teachers' teaching quality, set the teaching quality evaluation qualification threshold, and provide improvement suggestions and training plans for teachers whose scores are below the threshold.

[0125] It is understandable that the intelligent teaching management solution of the present invention has many significant beneficial effects. First, data processing is efficient. Through advanced data preprocessing and mining technology, it can quickly analyze massive teaching data and provide timely support for management decisions. Secondly, intelligent scheduling is more reasonable. The optimization algorithm that comprehensively considers multiple factors ensures that courses and classroom arrangements are scientific and efficient, greatly improving the utilization rate of teaching resources. In addition, real-time and accuracy are given equal importance, and the solution can be adjusted according to real-time data to ensure that management decisions are always in line with reality. This solution comprehensively improves the efficiency and quality of teaching management, bringing great convenience to education management.

[0126] In a preferred embodiment, the present application further provides an electronic device, the electronic device comprising: A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method for intelligent management of teaching affairs is implemented. The computer device can be broadly a server, a terminal, or any other electronic device with necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected via a system bus. The processor of the computer device can be used to provide necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory can provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, the steps of the method of the present invention are performed.

[0127] The present invention may be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of an embodiment of the present invention to be executed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be performed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be performed by one or more computer devices or processors, and one or more other method steps / operations may be performed by one or more other computer devices or processors. One or more computer devices or processors may perform a single method step / operation, or perform two or more method steps / operations.

[0128] It will be appreciated by a person skilled in the art that the method steps of the present invention may be performed by instructing related hardware such as a computer device or a processor through a computer program, and the computer program may be stored in a non-temporary computer-readable storage medium, which causes the steps of the present invention to be performed when the computer program is executed. Depending on the circumstances, any reference to memory, storage, database, or other media herein may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0129] It should be noted here that the intelligent teaching management solution of the present invention has many significant beneficial effects. First, data processing is efficient. Through advanced data preprocessing and mining technology, it can quickly analyze massive teaching data and provide timely support for management decisions. Secondly, intelligent scheduling is more reasonable. The optimization algorithm that comprehensively considers multiple factors ensures that courses and classroom arrangements are scientific and efficient, greatly improving the utilization rate of teaching resources. In addition, real-time and accuracy are given equal importance, and the solution can be adjusted according to real-time data to ensure that management decisions are always in line with reality. This solution comprehensively improves the efficiency and quality of teaching management, bringing great convenience to education management.

[0130] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0131] The specific implementation of the present invention described above does not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for intelligent management of teaching affairs, characterized in that: The method comprises: S1. Data collection and preprocessing: Through the data collection module, the school's existing system interface interacts and the sensor collects data, cleans the collected data, unifies the format and standardizes it, and performs feature extraction and feature engineering; S2. Student course recommendation: Based on students’ historical learning data and course information, a collaborative filtering algorithm and a content-based recommendation algorithm are combined to build a student course recommendation model, set a recommendation score threshold, and recommend courses with scores higher than the threshold to students; S3, Course Arrangement and Classroom Scheduling: Construct a course arrangement and classroom scheduling model, use a combination of genetic algorithm and simulated annealing algorithm to solve it, set the classroom capacity utilization threshold, and adjust the classroom allocation plan according to the threshold; S4. Teaching evaluation and quality monitoring: Design a teaching evaluation index system, collect multi-party evaluation data, use hierarchical analysis method and fuzzy comprehensive evaluation method to calculate the comprehensive score of teachers' teaching quality, set the qualified threshold for teaching quality evaluation, and provide improvement suggestions and training plans for teachers whose scores are below the threshold.

2. The intelligent teaching management method according to claim 1, characterized in that: In the data collection and preprocessing steps, the collected data includes student information, teacher information and course information. The student information includes student number, name, major, and grade. The teacher information includes work number, name, title, and teaching subjects. The course information includes course number, course name, credits, class time, and class location.

3. The intelligent teaching management method according to claim 1, characterized in that: In the student course recommendation step, the collaborative filtering algorithm calculates the similarity between students and makes recommendations based on the courses selected by similar students, and the content-based recommendation algorithm makes matching recommendations by analyzing the course content and the students' majors and interests.

4. The intelligent teaching management method according to claim 3 is characterized in that: The specific methods for students to recommend courses include: S21. Course content feature extraction: Preprocess the text information of the syllabus, textbook content, and course introduction of each course, including word segmentation, stop word removal, and stem extraction operations; convert the preprocessed text into a vector representation using a word vector model, and calculate the vector value of each word. The word vector model includes Word2Vec or GloVe; obtain the feature vector of the course content by calculating the average value and weighted average value of the word vectors in the text, which can reflect the core knowledge and theme of the course; S22. Student interest modeling: Collect students’ behavior data during the learning process, including course browsing records, course selection history, study time, and homework completion status; Based on students’ behavior data, use association rule mining or cluster analysis methods to discover students’ interest preferences and potential needs; Express students’ interest preferences in vector form, which has the same dimension as the course content feature vector; S23, matching recommendation calculation: Calculate the similarity between the student interest vector and the course content feature vector, using the cosine similarity or Euclidean distance similarity measurement method; sort the courses according to the similarity score, and select the courses with similarity higher than the recommended score threshold as recommended courses.

5. The intelligent teaching management method according to claim 4 is characterized in that: In the course arrangement and classroom scheduling step, the course time requirement, classroom capacity, multimedia equipment requirements, and teacher and student free time are input into the course arrangement and classroom scheduling model as constraints.

6. The intelligent teaching management method according to claim 5, characterized in that: In the course arrangement and classroom scheduling steps, the neighborhood search process of the simulated annealing operation includes: Validity check and correction of neighborhood solutions: After generating a neighborhood solution, check the validity of the neighborhood solution to ensure that the new course arrangement and classroom allocation plan meet all constraints, including that the course time does not conflict and the classroom capacity meets the student number requirements. For neighborhood solutions that do not meet the constraints, take corresponding correction measures, including: if there is a course time conflict, reselect a suitable classroom to eliminate the conflict; if the classroom capacity is insufficient, replace it with a classroom that can accommodate the corresponding number of students.

7. The intelligent teaching management method according to claim 5, characterized in that: In the course arrangement and classroom scheduling steps, the crossover operation of the genetic algorithm adopts a multi-point crossover method, which specifically includes the following steps: Select crossover points: for the two selected parent chromosomes, randomly select k crossover points, where the value range of k is [2, chromosome length / 4], and the chromosome length represents the number of courses; Gene exchange: exchange genes of two parent chromosomes at the selected k crossover points to generate two new daughter chromosomes; Conflict handling: Check whether there is a conflict between courses and classrooms in the newly generated offspring chromosomes. If there is a conflict, a greedy algorithm is used to adjust it, that is, giving priority to meeting the course schedule and classroom capacity constraints, and exchanging or adjusting the classroom allocation of some courses until the conflict is eliminated.

8. The intelligent teaching management method according to claim 1, characterized in that: In the teaching evaluation and quality monitoring steps, the teaching evaluation index system includes teaching content, teaching methods, and teacher teaching attitude.

9. An intelligent teaching management system, characterized in that: include: Data acquisition and preprocessing module: used to interact with the school's existing system interface and collect data from sensors through the data acquisition module, clean the collected data, unify the format and standardize it, and perform feature extraction and feature engineering; Student course recommendation module: Based on students' historical learning data and course information, a student course recommendation model is constructed by combining collaborative filtering algorithm and content-based recommendation algorithm, and a recommendation score threshold is set to recommend courses with scores higher than the threshold to students; Course arrangement and classroom scheduling module: build a course arrangement and classroom scheduling model, comprehensively consider multiple factors, use a combination of genetic algorithm and simulated annealing algorithm to solve, set the classroom capacity utilization threshold, and adjust the classroom allocation plan according to the threshold; Teaching evaluation and quality monitoring module: Design a teaching evaluation index system, collect multi-party evaluation data, use hierarchical analysis method and fuzzy comprehensive evaluation method to calculate the comprehensive score of teachers' teaching quality, set the qualified threshold for teaching quality evaluation, and provide improvement suggestions and training plans for teachers whose scores are below the threshold.

10. An electronic device, characterized in that: include: Memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the intelligent teaching management method according to any one of claims 1 to 8 is implemented.

Images