Education auxiliary system based on artificial intelligence

Abstract

The invention relates to the technical field of man-machine interaction, in particular to an education auxiliary system based on artificial intelligence. The system comprises a learning resource management unit which is used for collecting and managing teaching resources and student scores; a student evaluation unit which evaluates the learning level of the student through a K-Means clustering algorithm according to the student score of the learning resource management unit; the personalized recommendation unit is used for recommending proper learning materials and paths to the students through a mixed recommendation algorithm according to the evaluation result of the student evaluation unit; and the support service unit is used for collecting opinions and suggestions of the students according to a recommendation result of the personalized recommendation unit, and improving the service quality and content in time. And the advantages and weak links of the students are identified through deep analysis of test results. And through a hybrid recommendation algorithm, the recommendation accuracy and diversity are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of human-computer interaction, and in particular to an education assistance system based on artificial intelligence. Background Art

[0002] The AI-based educational assistance system is an intelligent platform that integrates a variety of advanced information technologies. It analyzes students' learning behaviors, abilities, and preferences to provide personalized learning plans and resource recommendations, helping students master knowledge and skills more efficiently. The system uses machine learning algorithms to predict learning outcomes, natural language processing technology to achieve intelligent dialogue and homework correction, data mining technology to discover learning patterns and potential problems, and a recommendation system to accurately push learning materials based on student characteristics. In addition, the system also supports teachers in classroom management and student evaluation, providing comprehensive teaching support. Through the comprehensive application of these technologies, the AI-based educational assistance system not only improves learning efficiency and teaching quality, but also promotes educational equity and personalized development, bringing profound changes to the field of education.

[0003] Among the existing AI-based educational assistance systems, the current AI education system has deficiencies in top-level design and lacks a clear definition of the core concepts of AI, leading to confusion in educational content and teaching methods. Teacher training is also a major challenge, as many teachers lack sufficient technical background and professional training to effectively utilize these advanced educational tools. Finally, although AI systems can provide personalized learning experiences, they are limited in terms of interpersonal communication and socialization skills training, and over-reliance on technology may affect students' interpersonal communication skills and teamwork spirit. Therefore, an AI-based educational assistance system is designed. Summary of the invention

[0004] The purpose of the present invention is to provide an educational assistance system based on artificial intelligence to solve the problem raised in the above background technology that in the existing educational assistance systems based on artificial intelligence, the current artificial intelligence education system has deficiencies in top-level design and lacks a clear definition of the core concepts of artificial intelligence, resulting in confusion in educational content and teaching methods. Teacher training is also a major challenge. Many teachers lack sufficient technical background and professional training and find it difficult to effectively use these advanced educational tools. Finally, although the AI ​​system can provide a personalized learning experience, it has limitations in interpersonal communication and socialization skills training. Over-reliance on technology may affect students' interpersonal communication skills and teamwork spirit.

[0005] To achieve the above object, the present invention aims to provide an artificial intelligence-based education assistance system, including a learning resource management unit, wherein the learning resource management unit is used to collect and manage teaching resources and student scores; A student evaluation unit, wherein the student evaluation unit evaluates the student's learning level according to the student's score of the learning resource management unit by using a K-Means clustering algorithm; A personalized recommendation unit, wherein the personalized recommendation unit recommends appropriate learning materials and paths to students through a hybrid recommendation algorithm according to the evaluation results of the student evaluation unit; The support service unit collects students' opinions and suggestions based on the recommendation results of the personalized recommendation unit and improves the service quality and content in a timely manner.

[0006] As a further improvement of the technical solution, the learning resource management unit includes a resource management module and a score collection module; Wherein, the resource management module is used to organize and classify learning resources; The score collection module is used to collect and record student scores.

[0007] As a further improvement of the technical solution, the student evaluation unit evaluates the student's learning level by using a K-Means clustering algorithm. The specific steps of evaluating the student's learning level are: S3.1. Obtain students' grades, homework scores, and test scores from the learning resource management unit; S3.2. Use the dropna() method of Pandas to remove invalid values, use the Z-score method to remove outliers, and use Z-score standardization to standardize the data, and use Min-Max normalization to normalize the data; S3.3. Use K-Means clustering algorithm to identify students’ strengths and weaknesses; S3.4. Generate student performance reports for reference by teachers and students.

[0008] As a further improvement of the technical solution, in S3.3, the specific steps of identifying the strengths and weaknesses of students by using the K-Means clustering algorithm are: S3.31, randomly select K samples as the initial centroids; S3.32, assign each sample to the cluster where the nearest centroid is located; S3.33, recalculate the center position of each cluster; S3.34, repeat the above two steps until the centroid no longer changes or the predetermined number of iterations is reached; S3.35. Use the silhouette coefficient to evaluate the quality of clustering results; S3.36. Observe the characteristics of each cluster and determine that the comprehensive development cluster represents the students' strengths, while the low performance cluster represents their weaknesses.

[0009] As a further improvement of the technical solution, in the personalized recommendation unit, the specific steps of recommending appropriate learning materials and paths to students through a hybrid recommendation algorithm are as follows: S4.1. Collect students’ personal information, learning history, interest preferences, and learning resources; S4.2. Use the dropna() method of Pandas to remove invalid values, use the Z-score method to remove outliers, and use Z-score standardization to standardize the data, and use Min-Max normalization to normalize the data; S4.3, extract features from user data, and combine the extracted features into a user profile; Among them, the features extracted from user data include learning style, interest preference, and learning progress; S4.4, calculating the similarity and recommending learning resources to the user based on the similarity; S4.5. Extract features from learning resources, match corresponding learning resources according to the interest preferences in the user profile, and recommend learning resources to users based on the matching degree; Among them, the features extracted from learning resources include topics, difficulty, and tags; S4.6. Build a training data set by extracting features from user data and learning resources, and use LSTM to train a recommendation model, and use the trained model to generate a recommendation list for the user; S4.7. Based on the expert experience method, assign a weight to each recommendation method and generate a final recommendation list.

[0010] As a further improvement of the technical solution, in S4.4, the similarity is calculated, and the specific expression for recommending learning resources to the user based on the similarity is: User similarity: ; in, Indicates Personal information of students; Indicates Personal information of students; Indicates students and students The similarity between Indicates The feature vector of students; Indicates The feature vector of students; Item Similarity: ; in, Indicates information about learning resources; Indicates information about learning resources; Representation Resources and resources The similarity between Generate a recommendation list: ; in, Indicates that collaborative filtering algorithm is used for students A list of recommended learning resources; Representation with students The most similar set of sets; Indicates students About Learning Resources Ratings and reviews.

[0011] As a further improvement of the technical solution, in S4.5, features are extracted from learning resources, and corresponding learning resources are matched according to the interest preferences in the user portrait. The specific expression for recommending learning resources to the user according to the matching degree is: Resource feature vector: ; in, Indicates The feature vector of a learning resource; express Features User interest matching: ; in, Indicates the User portraits of students; Indicates students portrait and resources The matching degree; Generate a recommendation list: ; in, Indicates that students are A list of recommended learning resources.

[0012] As a further improvement of the technical solution, in S4.6, a training data set is constructed by extracting features from user data and from learning resources, and a recommendation model is trained using LSTM, and a recommendation list is generated for the user using the trained model; Training dataset: ; in, represents the training data set; Indicates students About Resources The actual ratings and clicks of Model training: in, Indicates labels of students; Represents the model for prediction of the next interaction of students; Indicates the number of categories; The student predicted by the model About Resources Rating and click probability; Generate a recommendation list: ; in, It means that students can learn through deep learning methods. A list of recommended learning resources.

[0013] As a further improvement of the technical solution, in S4.7, a weight is assigned to each recommendation method according to the expert experience method, and a final recommendation list is generated; ; in, Indicates that the final student The comprehensive recommended learning resource list is obtained by weighted average of the results of various recommendation methods; Represents the weight of the collaborative filtering recommendation method; Represents the weight of the content filtering recommendation method; Represents the weight of the deep learning recommendation method.

[0014] As a further improvement of the technical solution, the support service unit includes a feedback collection module and an analysis and improvement module; Wherein, the feedback collection module is used to collect feedback from students after using personalized recommendations; The analysis and improvement module is used to analyze feedback and improve the system according to the analysis results.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This AI-based educational assistance system makes resources easy to find and use through classification and labeling, improving the accessibility and utilization of resources. Through various forms of tests, students' learning outcomes are comprehensively evaluated. Through in-depth analysis of test results, students' strengths and weaknesses are identified, providing a basis for personalized teaching.

[0016] 2. This AI-based education assistance system improves the accuracy and diversity of recommendations through a hybrid recommendation algorithm. It improves students' learning outcomes through rich resources, comprehensive evaluations, and personalized recommendations. It continuously optimizes service quality and content by collecting user feedback and suggestions, ensuring system stability and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is the overall flow chart of the present invention; The meaning of each number in the figure is: 1. Learning resource management unit; 2. Student assessment unit; 3. Personalized recommendation unit; 4. Support service unit. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] See also Figure 1 As shown, an artificial intelligence-based education assistance system is provided, including a learning resource management unit 1, which is used to collect and manage teaching resources and student scores; In this example, the learning resource management unit 1 includes a resource management module and a score collection module; Wherein, the resource management module is used to organize and classify learning resources; The score collection module is used to collect and record student scores.

[0020] The artificial intelligence-based education assistance system also includes a student evaluation unit 2, which evaluates the student's learning level through a K-Means clustering algorithm based on the student's score in the learning resource management unit 1; In this example, the student evaluation unit 2 evaluates the student's learning level by using the K-Means clustering algorithm. The specific steps of evaluating the student's learning level are: S3.1. Obtain the student's grades, homework scores, and test scores from the learning resource management unit (1); S3.2. Use the dropna() method of Pandas to remove invalid values, use the Z-score method to remove outliers, and use Z-score standardization to standardize the data, and use Min-Max normalization to normalize the data; S3.3. Use K-Means clustering algorithm to identify students’ strengths and weaknesses; In this example, the specific steps to identify students' strengths and weaknesses through the K-Means clustering algorithm are: S3.31, randomly select K samples as the initial centroids; S3.32, assign each sample to the cluster where the nearest centroid is located; S3.33, recalculate the center position of each cluster; S3.34, repeat the above two steps until the centroid no longer changes or the predetermined number of iterations is reached; S3.35. Use the silhouette coefficient to evaluate the quality of clustering results; S3.36. Observe the characteristics of each cluster and determine that the comprehensive development cluster represents the students' strengths, while the low performance cluster represents their weaknesses.

[0021] Among them, the comprehensive development cluster indicates that students' grades are relatively balanced, without obvious subject bias; the low performance cluster indicates that students' overall grades are low and they may need extra attention and support; Specifically, the implementation steps are: import pandas as pd from sklearn.cluster import KMeans from sklearn.preprocessing import StandardScaler from sklearn.metrics import silhouette_score import matplotlib.pyplot as plt # Assume that df is a DataFrame containing student score data df = pd.DataFrame({ 'student_id': [1, 2, 3, 4, 5, 6, 7, 8], 'math_score': [85, 92, 76, 89, 94, 83, 80, 79], 'english_score': [80, 85, 78, 90, 92, 85, 76, 79], 'science_score': [78, 88, 75, 87, 90, 80, 82, 76] }) # Standardize data scaler = StandardScaler() X_scaled = scaler.fit_transform(X) # Clustering using KMeans kmeans = KMeans(n_clusters=3, random_state=0) kmeans.fit(X_scaled) labels = kmeans.labels_ # Add cluster labels to the original data df['cluster'] = labels # Calculate the silhouette coefficient to assess the quality of the clustering silhouette_avg = silhouette_score(X_scaled, labels) print(f"Silhouette Score: {silhouette_avg}") # Visualize the results plt.figure(figsize=(10, 6)) plt.scatter(X_scaled[:, 0], X_scaled[:, 1], c=labels, s=50, cmap='viridis') centers = kmeans.cluster_centers_ plt.scatter(centers[:, 0], centers[:, 1], c='red', s=200, alpha=0.7) plt.title("Clusters of Students") plt.xlabel("Math Score") plt.ylabel("English Score") plt.show() # Output the label of each student print(df) S3.4. Generate student performance reports for reference by teachers and students.

[0022] Specifically, it can comprehensively and accurately evaluate students' learning situation, provide detailed scoring reports and feedback; identify students' strengths and weaknesses through quantitative analysis, and provide a scientific basis for personalized teaching; improve the efficiency and accuracy of evaluation, and reduce the workload of teachers; promote effective communication between teachers and students, and improve learning effects; continuously optimize the evaluation system to ensure the stability and reliability of the system. These advantages enable the system to better support students' learning and development, and improve the quality and efficiency of education.

[0023] The artificial intelligence-based education assistance system further includes a personalized recommendation unit 3, which recommends appropriate learning materials and paths for students through a hybrid recommendation algorithm based on the evaluation results of the student evaluation unit 2; In this example, the specific steps of recommending appropriate learning materials and paths for students through the hybrid recommendation algorithm are: S4.1. Collect students’ personal information, learning history, interest preferences, and learning resources; S4.2. Use the dropna() method of Pandas to remove invalid values, use the Z-score method to remove outliers, and use Z-score standardization to standardize the data, and use Min-Max normalization to normalize the data; Specifically, the implementation steps are: import pandas as pd from scipy import stats from sklearn.preprocessing import StandardScaler, MinMaxScaler # Load data data = pd.read_csv('student_data.csv') # Remove invalid values data_cleaned = data.dropna() # Remove outliers (using Z-score method) data_cleaned = data_cleaned[(np.abs(stats.zscore(data_cleaned)) < 3).all(axis=1)] # Data standardization (Z-score standardization) scaler = StandardScaler() data_standardized = scaler.fit_transform(data_cleaned) # Data normalization (Min-Max normalization) min_max_scaler = MinMaxScaler() data_normalized = min_max_scaler.fit_transform(data_cleaned) S4.3, extract features from user data, and combine the extracted features into a user profile; Among them, the features extracted from user data include learning style, interest preference, and learning progress; Specifically, the steps for extracting features from user data and combining the extracted features into a user profile are as follows: # Example User Data user_data = pd.read_csv('user_data.csv') # Extract learning style learning_styles = user_data['Learning Style'] # Extract interest preferences interests = user_data['Interests'].apply(lambda x: x.split(',')) # Extract learning progress progress = user_data[['Exam Scores', 'Homework Scores', 'ClassPerformance']] # Combine into user portrait user_profile=pd.concat([learning_styles,interests.apply(pd.Series),progress], axis=1) user_profile.columns=['Learning Style',Interest1','Interest2', 'ExamScores', 'Homework Scores', 'Class Performance'] S4.4, calculating the similarity and recommending learning resources to the user based on the similarity; In this example, the similarity is calculated and the specific expression for recommending learning resources to users based on the similarity is: User similarity: ; in, Indicates Personal information of students; Indicates Personal information of students; Indicates students and students The similarity between Indicates The feature vector of students; Indicates The feature vector of students; Item Similarity: ; in, Indicates Information about learning resources; Indicates information about learning resources; Representation Resources and resources The similarity between Generate a recommendation list: ; in, Indicates that collaborative filtering algorithm is used for students A list of recommended learning resources; Representation with students The most similar set of sets; Indicates students About Learning Resources Ratings and reviews.

[0024] S4.5. Extract features from learning resources, match corresponding learning resources according to the interest preferences in the user profile, and recommend learning resources to users based on the matching degree; Among them, the features extracted from learning resources include topics, difficulty, and tags; Specifically, the steps for extracting features from learning resources are: # Sample learning resource data resource_data = pd.read_csv('resource_data.csv') # Extract the subject topics = resource_data['Topic'] # Extraction difficulty difficulties = resource_data['Difficulty'] # Extract tags tags = resource_data['Tags'].apply(lambda x: x.split(',')) # Combine into learning resource features resource_features=pd.concat([topics,difficulties,tags.apply(pd.Series)], axis=1) resource_features.columns = ['Topic', 'Difficulty', 'Tag1', 'Tag2'] In this example, features are extracted from learning resources, and corresponding learning resources are matched according to the interest preferences in the user profile. The specific expression for recommending learning resources to users based on the matching degree is: Resource feature vector: ; in, Indicates The feature vector of a learning resource; express Features User interest matching: ; in, Indicates the User portraits of students; Indicates students portrait and resources The matching degree; Generate a recommendation list: ; in, Indicates that students are A list of recommended learning resources.

[0025] S4.6. Build a training data set by extracting features from user data and learning resources, and use LSTM to train a recommendation model, and use the trained model to generate a recommendation list for the user; In this example, features are extracted from user data and learning resources to construct a training data set, and a recommendation model is trained using LSTM. The trained model is used to generate a recommendation list for the user. Training dataset: ; in, represents the training data set; Indicates students About Resources The actual ratings and clicks of Model training: in, Indicates labels of students; Represents the model for prediction of the next interaction of students; Indicates the number of categories; The student predicted by the model About Resources Rating and click probability; Generate a recommendation list: ; in, It means that students can learn through deep learning methods. A list of recommended learning resources.

[0026] S4.7. Based on the expert experience method, assign a weight to each recommendation method and generate a final recommendation list.

[0027] In this example, a weight is assigned to each recommendation method based on the expert experience method, and a final recommendation list is generated; ; in, Indicates that the final student The comprehensive recommended learning resource list is obtained by weighted average of the results of various recommendation methods; Represents the weight of the collaborative filtering recommendation method, which assigns weights based on the similarity between users; Represents the weight of the content filtering recommendation method, which is assigned based on the matching degree between user profile and resource features; Represents the weight of the deep learning recommendation method, which assigns weights based on the predictions of the deep learning model.

[0028] Specifically, it can comprehensively collect and process user data to ensure the accuracy and personalization of recommendations; improve the diversity and accuracy of recommendations through the comprehensive application of multiple recommendation methods; use LSTM models to process time series data to capture users' learning dynamics and preference changes; collect students' feedback and suggestions in a timely manner to continuously optimize the recommendation system and service quality; promote interactive communication between teachers and students, and improve learning outcomes and satisfaction. These advantages enable the system to better support students' learning and development and improve the quality and efficiency of education.

[0029] The artificial intelligence-based education assistance system also includes a support service unit 4, which collects students' opinions and suggestions based on the recommendation results of the personalized recommendation unit 3 and improves the service quality and content in a timely manner.

[0030] In this example, the support service unit (4) includes a feedback collection module and an analysis and improvement module; Wherein, the feedback collection module is used to collect feedback from students after using personalized recommendations; The analysis and improvement module is used to analyze feedback and improve the system according to the analysis results.

[0031] Specifically, it can obtain students' feedback in a timely manner, understand users' real needs and experiences; optimize platform functions and services through data analysis, improve system stability and reliability; provide timely technical support to help students solve problems during use, improve user experience and satisfaction; promote interactive communication between teachers and students, and enhance the cohesion and activity of the learning community. These advantages enable the system to better adapt to user needs and continuously improve the quality and efficiency of education.

[0032] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. An artificial intelligence-based educational assistance system, characterized in that: include: A learning resource management unit (1), the learning resource management unit (1) is used to collect and manage teaching resources and student scores; A student evaluation unit (2), wherein the student evaluation unit (2) evaluates the student's learning level according to the student's performance in the learning resource management unit (1) by using a K-Means clustering algorithm; A personalized recommendation unit (3), wherein the personalized recommendation unit (3) recommends appropriate learning materials and paths to the student through a hybrid recommendation algorithm according to the evaluation result of the student evaluation unit (2); The support service unit (4) collects students' opinions and suggestions based on the recommendation results of the personalized recommendation unit (3) and improves the service quality and content in a timely manner.

2. The artificial intelligence-based education assistance system according to claim 1, characterized in that: The learning resource management unit (1) comprises a resource management module and a score collection module; Wherein, the resource management module is used to organize and classify learning resources; The score collection module is used to collect and record student scores.

3. The artificial intelligence-based education assistance system according to claim 1, characterized in that: The student evaluation unit (2) evaluates the student's learning level by using the K-Means clustering algorithm. The specific steps of evaluating the student's learning level are: S3.

1. Obtain the student's grades, homework scores, and test scores from the learning resource management unit (1); S3.

2. Use the dropna() method of Pandas to remove invalid values, use the Z-score method to remove outliers, and use Z-score standardization to standardize the data, and use Min-Max normalization to normalize the data; S3.

3. Use K-Means clustering algorithm to identify students’ strengths and weaknesses; S3.

4. Generate student performance reports for reference by teachers and students.

4. The artificial intelligence-based education assistance system according to claim 3, characterized in that: In S3.3, the specific steps of identifying students' strengths and weaknesses through the K-Means clustering algorithm are: S3.31, randomly select K samples as the initial centroids; S3.32, assign each sample to the cluster where the nearest centroid is located; S3.33, recalculate the center position of each cluster; S3.34, repeat the above two steps until the centroid no longer changes or the predetermined number of iterations is reached; S3.

35. Use the silhouette coefficient to evaluate the quality of clustering results; S3.

36. Observe the characteristics of each cluster and determine that the comprehensive development cluster represents the students' strengths, while the low performance cluster represents their weaknesses.

5. The artificial intelligence-based education assistance system according to claim 1, characterized in that: In the personalized recommendation unit (3), the specific steps of recommending appropriate learning materials and paths to students through the hybrid recommendation algorithm are as follows: S4.

1. Collect students’ personal information, learning history, interest preferences, and learning resources; S4.

2. Use the dropna() method of Pandas to remove invalid values, use the Z-score method to remove outliers, and use Z-score standardization to standardize the data, and use Min-Max normalization to normalize the data; S4.3, extract features from user data, and combine the extracted features into a user profile; Among them, the features extracted from user data include learning style, interest preference, and learning progress; S4.4, calculating the similarity and recommending learning resources to the user based on the similarity; S4.

5. Extract features from learning resources, match corresponding learning resources according to the interest preferences in the user profile, and recommend learning resources to users based on the matching degree; Among them, the features extracted from learning resources include topics, difficulty, and tags; S4.

6. Build a training data set by extracting features from user data and learning resources, and use LSTM to train a recommendation model, and use the trained model to generate a recommendation list for the user; S4.

7. Based on the expert experience method, assign a weight to each recommendation method and generate a final recommendation list.

6. The artificial intelligence-based education assistance system according to claim 5, characterized in that: In S4.4, similarity is calculated, and the specific expression for recommending learning resources to the user based on the similarity is: User similarity: ； in, Indicates Personal information of students; Indicates Personal information of students; Indicates students and students The similarity between Indicates The feature vector of students; Indicates The feature vector of students; Item Similarity: ； in, Indicates information about learning resources; Indicates information about learning resources; Representation Resources and resources The similarity between Generate a recommendation list: ； in, Indicates that collaborative filtering algorithm is used for students A list of recommended learning resources; Representation with students The most similar set of sets; Indicates students About Learning Resources Ratings and reviews.

7. The artificial intelligence-based education assistance system according to claim 5, characterized in that: In S4.5, features are extracted from learning resources, and corresponding learning resources are matched according to the interest preferences in the user profile. The specific expression for recommending learning resources to the user according to the matching degree is: Resource feature vector: ； in, Indicates The feature vector of a learning resource; express Features User interest matching: ； in, Indicates the User portraits of students; Indicates students portrait and resources The matching degree; Generate a recommendation list: ； in, Indicates that students are A list of recommended learning resources.

8. The artificial intelligence-based education assistance system according to claim 5, characterized in that: In S4.6, a training data set is constructed by extracting features from user data and from learning resources, and a recommendation model is trained using LSTM, and a recommendation list is generated for the user using the trained model; Training dataset: ； in, represents the training data set; Indicates students About Resources The actual ratings and clicks of Model training: in, Indicates labels of students; Represents the model for prediction of the next interaction of students; Indicates the number of categories; The student predicted by the model About Resources Rating and click probability; Generate a recommendation list: ； in, It means that students can learn through deep learning methods. A list of recommended learning resources.

9. The artificial intelligence-based education assistance system according to claim 5, characterized in that: In S4.7, a weight is assigned to each recommendation method according to the expert experience method, and a final recommendation list is generated; ； in, Indicates that the final student The comprehensive recommended learning resource list is obtained by weighted average of the results of various recommendation methods; Represents the weight of the collaborative filtering recommendation method; Represents the weight of the content filtering recommendation method; Represents the weight of the deep learning recommendation method.

10. The artificial intelligence-based education assistance system according to claim 1, characterized in that: The support service unit (4) includes a feedback collection module and an analysis and improvement module; Wherein, the feedback collection module is used to collect feedback from students after using personalized recommendations; The analysis and improvement module is used to analyze feedback and improve the system according to the analysis results.

Images