Personalized course recommendation method and system based on knowledge graph

Abstract

The invention discloses a personalized course recommendation method and system based on a knowledge graph, and relates to the technical field of course recommendation. The personalized course recommendation method and system based on the knowledge graph comprises the following steps: S1, acquiring user learning data by collecting front-end burying points and background access log data of a learning platform in real time, and preprocessing the user learning data; s2, constructing a behavior data relationship between the user and the knowledge points, evaluating a mastering condition of the user on the knowledge points, and constructing a time sequence knowledge point score set; s3, based on the time sequence knowledge point score set, identifying abnormally mastered knowledge points of the user in the current stage, analyzing whether the courses are worthy of recommendation, and partitioning candidate courses and generating recommended courses; and S4, quantifying the response difference between the recommendation course and the mastering effect, updating the recommendation tag and the user record, and perfecting the recommendation data circulation. The problem that in the prior art, learning effects are difficult to measure, and behavior-driven recommendation circulation is likely to happen is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of course recommendation, and specifically provides a personalized course recommendation method and system based on a knowledge graph. Background Art

[0002] With the wide application of online education platforms, course recommendation systems for users play an important role in improving learning efficiency and learning experience. Currently, most recommendation systems mainly rely on behavioral data such as user click records, browsing frequencies, course collections, and learning completion rates for modeling, and generate personalized recommendation results through technical means such as collaborative filtering, content matching, or knowledge graph reasoning.

[0003] For example, the invention with the publication number: CN116542731A discloses a fitness plan course recommendation method based on a knowledge graph, which relates to the field of fitness personalized recommendation. The fitness plan course recommendation method based on a knowledge graph includes collecting user physical fitness characteristic information and a fitness course action data set, extracting knowledge attributes from the fitness course action descriptions, structuring the knowledge graph, taking a user physical fitness characteristic u and a course v as inputs, and the output is the probability that the user physical fitness characteristic u can perform the training course action v, and adjusting the hyperparameters of the model.

[0004] For example, the invention with the publication number: CN116501970A provides an online course recommendation method based on a knowledge graph and convolution. The model first uses a feature extraction module to extract and transform the user's historical interaction information and the feature information in the course field into embedding vectors, and fuses these vectors into the knowledge graph to generate embedding expressions of the user and the course. Finally, through a message propagation algorithm, this information is passed to adjacent nodes to obtain the final embedding vectors of the user and the course for recommendation.

[0005] However, the learning behavior itself has a high degree of goal orientation and stage progression. Simply relying on behavioral indicators such as click-through rate or completion rate is difficult to accurately reflect the learner's true mastery of knowledge content. On the other hand, most existing recommendation systems take user interest as the core driving goal and lack a feedback mechanism based on ability improvement and learning effectiveness, which easily leads to the recommendation system falling into a "behavior-driven recommendation loop".

[0006] Therefore, in view of the above problems, there is an urgent need for a personalized course recommendation method and system based on a knowledge graph. Summary of the Invention

[0007] Technical Problem to be Solved Aiming at the deficiencies of the prior art, the present invention provides a personalized course recommendation method and system based on a knowledge graph, which solves the problems that the prior art is difficult to measure learning effects and easily falls into a behavior-driven recommendation loop.

[0008] Technical solution: To achieve the above objectives, the present invention is realized through the following technical solutions: A personalized course recommendation method based on a knowledge graph, comprising the following steps: S1, obtaining user learning data by collecting real-time front-end buried points and back-end access log data of the learning platform, and preprocessing the user learning data to obtain preprocessed user learning data; S2, constructing the behavioral data relationship between users and knowledge points, normalizing and splitting the preprocessed user learning data, evaluating the user's mastery of knowledge points, and constructing a time series knowledge point scoring set; S3, identifying the abnormally mastered knowledge points of the user at the current stage based on the time series knowledge point scoring set, conducting differential analysis in combination with the behavioral data relationship, evaluating the recommendation value of candidate courses, partitioning the candidate courses, and generating recommended courses; S4, quantifying the response difference between the recommended courses and the mastery effectiveness from the ratio dimension, updating the recommendation tags and user records, and improving the recommendation data loop.

[0009] Further, the specific steps for obtaining user learning data by collecting real-time front-end buried points and back-end access log data of the learning platform are as follows: Obtaining user learning data by collecting real-time front-end buried points and back-end access log data of the learning platform, where the user learning data includes course code, course browsing duration, course access times, historical average course browsing duration, video viewing completeness, course scoring data, course scoring full marks data, course test score data, course test full marks data, and the number of knowledge points involved in the course; among them, the course code is extracted through the course name field in the course access log; the course browsing duration is obtained by recording the difference between the start time and the exit time when the user enters the course page each time and summing up by course; the course access times are obtained by counting the total number of times the user accesses a certain course page; the historical average course browsing duration is obtained by dividing the total browsing duration of the user for the same course within a week by the course access times; the video viewing completeness is obtained by calculating the ratio of the actual playing duration recorded by the video player to the total duration of the course video; the course scoring data and the course scoring full marks data are obtained by extracting the scoring field submitted by the user after completing the course learning; the course test score data and the course test full marks data are obtained by recording the scoring results of the user participating in the in-course quiz after completing the course; the number of knowledge points involved in the course is obtained by analyzing the mapping relationship between the course and the knowledge points in the course resource management database and counting the number of knowledge point tags covered by each course.

[0010] Furthermore, the specific steps for preprocessing the user learning data to obtain the preprocessed user learning data are as follows: Combine the median absolute deviation method of a sliding time window with the isolation forest algorithm to perform local and global anomaly detection on the user learning data, and eliminate abnormal user learning data including abnormal long-term hanging up and skipping classes to brush progress; Use the similar user collaborative completion strategy and the exponentially weighted moving average algorithm to fill in the missing data fields of the user learning data caused by system delays, missing data points, or terminal fluctuations; Fit and denoise the user learning data using the Bayesian dynamic smoothing method and the locally weighted regression filtering algorithm; Transform and compress the range of the user learning data through exponential transformation combined with the maximum-minimum normalization method to achieve normalization processing.

[0011] Furthermore, construct the behavioral data relationship between users and knowledge points, and perform normalized splitting on the preprocessed user learning data. The specific steps for evaluating the user's mastery of knowledge points are as follows: According to the knowledge point labels marked in each course, extract the corresponding knowledge point set; Combine the course codes in the user learning records, and search one by one for the knowledge point labels included in the courses the user has learned, and merge them at the user dimension to obtain the entire knowledge point set learned by the user; Use the user name and knowledge point labels as corresponding items to organize the association records between users and knowledge points, and form the behavioral correspondence relationship between users and knowledge points during the learning process; Based on the established behavioral correspondence relationship between users and knowledge points, for each knowledge point, evaluate the user's mastery of this knowledge point by integrating the preprocessed user learning data such as the course browsing duration, video viewing completeness, and course rating data generated by the user in the courses containing the same knowledge point: For a course containing this knowledge point, calculate the ratio between the course browsing duration and the historical average browsing duration of the course, and multiply this ratio by the browsing duration weight, and the resulting product is used as the browsing duration score; Multiply the video viewing completeness by the video completeness weight, and the resulting product is used as the video completeness score; Divide the course rating data by the full score data of the course rating, and then multiply the ratio by the course rating weight, and the resulting product is used as the course rating data score; Add the browsing duration score, the video completeness score, and the course rating data score, and then multiply by the reciprocal of the number of knowledge points involved in the course to obtain the apportioned score of the user for this knowledge point in the course; Recalculate the apportioned score for all courses containing this knowledge point, and sum the apportioned scores of each course to obtain the knowledge point mastery evaluation value.

[0012] Furthermore, the specific steps for constructing a time series knowledge point scoring set are as follows: after calculating the knowledge point mastery evaluation value, the knowledge point mastery evaluation value and the mastery threshold are compared in real time: when the knowledge point mastery evaluation value is greater than or equal to the mastery threshold, it is marked as a normally mastered knowledge point and no adjustment is required; when the knowledge point mastery evaluation value is less than the mastery threshold, it is marked as an abnormally mastered knowledge point, the user's current learning path is prompted, and a recommendation instruction for adding supplementary courses related to this knowledge point is generated; the compared knowledge point mastery evaluation value is time-serialized: the knowledge point mastery evaluation value generated by the user in each learning behavior is recorded in chronological order, and the corresponding timestamp, course code and knowledge point label are marked; all knowledge point mastery evaluation values of the same user on the same knowledge point are sorted in chronological order to construct a knowledge point mastery evaluation value sequence of the user on this knowledge point; the knowledge point mastery evaluation value sequence of the user on all knowledge points is aggregated and sorted to form a time series knowledge point mastery evaluation value set of the user under the entire knowledge structure.

[0013] Furthermore, based on the time series knowledge point scoring set, the abnormal mastery of knowledge points in the user's current stage is identified, and differentiation analysis is carried out in combination with the behavioral data relationship. The specific steps for evaluating the recommendation value of candidate courses are as follows: Combine the time series knowledge point mastery evaluation value set of the user under the entire knowledge structure, extract the set of knowledge points marked as abnormal mastery in the current period, and count the number of knowledge points marked as abnormal mastery; group the abnormal records according to the knowledge point labels, identify the knowledge point labels that are continuously marked as abnormal within a short period of time within the same user, and form the key supplementary knowledge point set of the user's current stage; based on the identified key supplementary knowledge point set, review all courses associated with each knowledge point, eliminate courses that the user has completed and courses that have not been watched completely, and include the remaining courses in the candidate course set; For each course in the candidate course set, calculate the course's historical average rating, extract the number of overlaps between the course's associated knowledge points and the set of supplementary knowledge points, and analyze whether the course is worthy of recommendation; calculate the square value of the number of overlaps between the course's associated knowledge points and the set of supplementary knowledge points, calculate the product of the number of knowledge points involved in the course and the number of knowledge points marked as abnormally mastered, divide this square value by this product, and use the ratio as the knowledge point matching score; calculate the course browsing time divided by the historical average browsing time of the course, and add this ratio to the video viewing completeness to obtain the learning input score; calculate the course rating data plus the historical average course rating, and divide the sum by twice the full score of the course rating data to obtain the course rating data performance score; multiply the knowledge point matching score, learning input score, and course rating data performance score to obtain the course recommendation strength assessment value.

[0014] Furthermore, the specific steps for partitioning the candidate courses and generating the recommended courses are as follows: Based on the course recommendation intensity evaluation value, construct a multi-dimensional course recommendation popularity partition according to the course recommendation threshold: Among them, the course recommendation threshold includes the first-level course recommendation threshold and the second-level course recommendation threshold; When the course recommendation intensity evaluation value is greater than or equal to the second-level course recommendation threshold, it is determined as a high-popularity course, directly included in the current recommended courses, and pushed with priority sorting; When the course recommendation intensity evaluation value is greater than the first-level course recommendation threshold and less than the second-level course recommendation threshold, it is determined as a medium-popularity course, enters the waiting area, aggregates themes through the supplementary knowledge point tags covered by the course, preferentially matches according to the user's historical unlearned content, and recommends 1 to 2 courses according to the tag distribution to expand the recommendation coverage; When the course recommendation intensity evaluation value is less than or equal to the first-level course recommendation threshold, it is determined as a low-popularity course and does not enter the current recommended courses; Generate the recommended courses for the user at the current stage according to the results of the multi-dimensional course recommendation popularity partition, and send the recommended courses to the user learning interface; Mark the supplementary knowledge point tags covered by each course in the recommended course list.

[0015] Furthermore, the specific steps for quantifying the response difference between the recommended courses and the mastery effect from the proportion dimension are as follows: After the user completes the study of the recommended courses, extract the course codes, course browsing durations, and supplementary knowledge point tags covered by the study of the current recommended courses, and establish the content structure of the recommended course study; Synchronously record the course test score data after the recommended courses, and update the corresponding knowledge point mastery evaluation value in the knowledge point dimension; Construct the comparison relationship of the knowledge point mastery evaluation values before and after the recommendation according to the user and the knowledge points; According to the comparison relationship of the knowledge point mastery evaluation values before and after the recommendation, quantify the response difference between the recommended courses and the mastery effect: Subtract the knowledge point mastery evaluation value of this knowledge point before the recommended course is issued from the knowledge point mastery evaluation value of this knowledge point after the recommended course is issued, then divide by the full score data of the course score, and then subtract the ratio of the course test score data to the full score data of the course test from this ratio. The absolute value of the difference is used as the mastery change ratio; Add 1 to the number of recommended courses involving this knowledge point in the recommended courses, then take the logarithm to the base 2, and add 1 to the result as the course coverage factor; Multiply the mastery change ratio and the course coverage factor to obtain the mastery feedback response value.

[0016] Furthermore, the specific steps to update the recommended tags and user records and improve the recommended data loop are as follows: After calculating the mastery feedback response value, compare the mastery feedback response value with the mastery feedback threshold in real time: Among them, the mastery feedback threshold includes the first-level mastery feedback threshold and the second-level mastery feedback threshold; When the mastery feedback response value is greater than or equal to the second-level mastery feedback threshold, the feedback is normal, and the knowledge points covered in the current course are marked as recommended valid tags, enhancing the recommended priority of the current course in the candidate course ranking for subsequent similar users; When the mastery feedback response value is greater than the first-level mastery feedback threshold and less than the second-level mastery feedback threshold, the feedback is lagging, and the current course is marked as an intermediate transition course and incorporated into the next recommendation cycle as the continuous observation object of the delayed supplementary content; When the mastery feedback response value is less than or equal to the first-level mastery feedback threshold, the feedback is abnormal, and the knowledge points covered in the current course are marked as recommended offset tags, marking the current course as a recommended insensitive course, and preferentially avoiding courses with similar teaching structures in the subsequent supplementary course screening; Arrange the mastery feedback response values generated by the same knowledge point in different recommendation cycles in sequence according to the user dimension to construct a mastery feedback response value sequence; For knowledge points with the mastery feedback response value less than the first-level mastery feedback threshold three times or more, extract the course code and course score data corresponding to the knowledge point, and recalculate the knowledge point mastery evaluation value of the user in this course based on the original course browsing duration, video viewing completeness, and course score data, replacing the original result for feedback correction; Write the course code, knowledge point tags, course browsing duration, course score data, video viewing completeness, mastery feedback response value, course test score data, and evaluation value change results generated by each round of recommended course learning into the user recommendation record; In the process of generating a new round of recommendations, preferentially read the user recommendation record, dynamically adjust the initial recommended course ranking, popularity differentiation strategy, and recommended threshold calculation logic, update the binding strength between the candidate course and the key knowledge point, and complete the active adaptive adjustment of the recommended path, so as to realize the recommendation self-closed-loop mechanism based on ability feedback.

[0017] The second aspect of the present invention provides a personalized course recommendation system based on a knowledge graph, including: a user learning data collection and preprocessing module, a knowledge point mastery evaluation module, a supplementary course recommendation generation module, and a recommendation effect feedback and correction module, where: the user learning data collection and preprocessing module is used to obtain user learning data by collecting real-time front-end buried point and background access log data of the learning platform, and preprocess the user learning data to obtain preprocessed user learning data; the knowledge point mastery evaluation module is used to construct the behavioral data relationship between the user and the knowledge points, perform normalized splitting on the preprocessed user learning data, evaluate the user's mastery of the knowledge points, and construct a time series knowledge point scoring set; the supplementary course recommendation generation module is used to identify the abnormally mastered knowledge points of the user at the current stage based on the time series knowledge point scoring set, conduct differential analysis in combination with the behavioral data relationship, evaluate the recommendation value of the candidate courses, partition the candidate courses, and generate recommended courses; the recommendation effect feedback and correction module is used to quantify the response difference between the recommended courses and the mastery results from the proportion dimension, update the recommendation tags and user records, and improve the recommendation data loop.

[0018] Beneficial effects The present invention has the following beneficial effects: (1) The personalized course recommendation method and system based on the knowledge graph propose a precise knowledge point mastery evaluation method by integrating user learning data such as course browsing duration, video viewing completeness, and course scoring data, making the evaluation of the knowledge point mastery status in the user's learning process more refined and accurate, and breaking through the limitation of the traditional system that only relies on click-through rate and completion rate to measure learning effectiveness.

[0019] (2) The personalized course recommendation method and system based on the knowledge graph dynamically adjust the feedback on learning effects by quantifying the difference between the knowledge point mastery evaluation values before and after recommendation and the course test score data, and combining the number of recommended courses and the user's behavioral responses, reflecting the real impact of learning behavior on ability improvement, and providing a real-time feedback basis for the optimization of recommended content.

[0020] (3) The personalized course recommendation method and system based on the knowledge graph designs a partition recommendation strategy for high, medium, and low popularity based on the course recommendation intensity evaluation, and combines the actual mastery feedback of the recommended content, and adopts a differential push strategy for courses with different intensities, balancing precise recommendation and exploratory recommendation, and improving the adaptability and coverage of the system.

[0021] (4) The personalized course recommendation method and system based on the knowledge graph realizes dynamic tracking and phased identification of the knowledge mastery status by constructing a time series mastery evaluation record between users and knowledge points, and constructs an abnormal mastery identification and supplementary content generation mechanism based on the knowledge point dimension, effectively improving the timeliness and accuracy of the recommendation decision.

[0022] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a flowchart of the personalized course recommendation method based on the knowledge graph; Figure 2 is a structural diagram of the personalized course recommendation system based on the knowledge graph; Figure 3 is a visualized bar chart of the knowledge point mastery evaluation value; Figure 4 is a line chart of the mastery evaluation value and the mastery feedback response value before and after the recommended course. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0025] Please refer to Figures 1-4 , the embodiments of the present invention provide a technical solution: a personalized course recommendation method based on the knowledge graph, including the following steps: S1, obtaining user learning data by real-time collecting the front-end buried point and background access log data of the learning platform, and preprocessing the user learning data to obtain the preprocessed user learning data; S2, constructing the behavioral data relationship between users and knowledge points, normalizing and splitting the preprocessed user learning data, evaluating the user's mastery of knowledge points, and constructing a time series knowledge point scoring set; S3, identifying the abnormally mastered knowledge points of the user at the current stage based on the time series knowledge point scoring set, conducting differential analysis in combination with the behavioral data relationship, evaluating the recommendation value of the candidate courses, partitioning the candidate courses and generating recommended courses; S4, quantifying the response difference between the recommended courses and the mastery effectiveness from the ratio dimension, updating the recommendation tags and user records, and improving the recommendation data loop.

[0026] Specifically, the specific steps to obtain user learning data by collecting front-end buried point and back-end access log data of the learning platform in real time are as follows: Obtain user learning data by collecting front-end buried point and back-end access log data of the learning platform in real time. User learning data includes course code, course browsing duration, course access times, historical average course browsing duration, video viewing completeness, course rating data, full marks of course rating data, course test score data, full marks of course test data, and the number of knowledge points involved in the course. Among them, the course code is extracted through the course name field in the course access log; the course browsing duration is obtained by recording the difference between the start time and the end time when the user enters the course page each time and summing up by course; the course access times are obtained by counting the total number of times the user accesses a certain course page; the historical average course browsing duration is obtained by dividing the total browsing duration of the user for the same course within a week by the course access times; the video viewing completeness is obtained by calculating the ratio of the actual playing duration recorded by the video player to the total duration of the course video; the course rating data and the full marks of course rating data are obtained by extracting the rating field submitted by the user after completing the course learning; the course test score data and the full marks of course test data are obtained by recording the rating results of the in-course quizzes participated by the user after completing the course; the number of knowledge points involved in the course is obtained by analyzing the mapping relationship between the course and the knowledge points in the course resource management database and counting the number of knowledge point labels covered by each course.

[0027] In this implementation plan, by precisely defining and collecting the course code, course browsing duration, course access times, historical average course browsing duration, video viewing completeness, course rating data, full marks of course rating data, course test score data, full marks of course test data, and the number of knowledge points involved in the course, a multi-dimensional data foundation covering the entire process of user behavior is constructed. This method effectively integrates front-end buried point behavior data and back-end access log information, and realizes a high-precision mapping of user learning behavior and learning content without changing the consistency of data names, providing accurate, stable and clearly structured data support for subsequent knowledge point mastery evaluation, course recommendation and feedback optimization, and breaking through the bottleneck of insufficient description of the user learning state by traditional recommendation systems.

[0028] Specifically, the user learning data is preprocessed to obtain the preprocessed user learning data in the following specific steps: by combining the median absolute deviation method of the sliding time window with the isolation forest algorithm, local and global anomaly detection is performed on the course browsing time, course visit times and course rating data in the user learning data, focusing on eliminating abnormal user learning data that violates the normal learning path rules, including abnormally long idle time and skipping classes to brush up progress; through the similar user collaborative completion strategy and the exponentially weighted moving average algorithm, the field missing problems of course rating data, course test score data and video viewing completeness caused by system delays, point loss and terminal fluctuations are intelligently completed to ensure the integrity and continuity of user learning data; the user learning data is fitted and denoised by the Bayesian dynamic smoothing method and the local weighted regression filtering algorithm to improve data stability; the user learning data is transformed and range compressed by the exponential transformation combined with the maximum and minimum normalization method to achieve normalization processing, providing a structural and consistent preprocessing data foundation for subsequent analysis.

[0029] In this implementation plan, by systematically preprocessing the course browsing time, course visit times, course rating data, course test score data and video viewing completeness in the user learning data, it is possible to effectively identify and eliminate abnormal user learning data, repair field missing problems, eliminate data noise, unify variable scales, and significantly improve the integrity, stability and comparability of the preprocessed user learning data. This provides a high-quality, structured data foundation for the subsequent construction of behavioral data relationships between users and knowledge points, evaluation of knowledge point mastery, and generation of recommended courses, ensuring the accuracy and robustness of the recommendation system.

[0030] Specifically, the specific steps for constructing the behavioral data relationship between users and knowledge points, normalizing and splitting the preprocessed user learning data, and evaluating the user's mastery of knowledge points are as follows: According to the knowledge point tags marked in each course, extract the corresponding knowledge point set; Combine the course codes in the user learning record, and search for the knowledge point tags included in the courses learned by the user one by one, and merge them at the user dimension to obtain the entire knowledge point set learned by the user; Using the user name and knowledge point tags as corresponding items, organize the association records between users and knowledge points to form the behavioral correspondence between users and knowledge points during the learning process; Based on the established behavioral correspondence between users and knowledge points, for each knowledge point, by integrating the preprocessed user learning data such as the course browsing duration, video viewing completeness, and course rating data generated by the user in the courses containing the same knowledge point, evaluate the user's mastery of this knowledge point: For a course containing this knowledge point, calculate the ratio between the course browsing duration and the historical average browsing duration of the course, and multiply this ratio by the browsing duration weight, and the resulting product is used as the browsing duration score; Multiply the video viewing completeness by the video completeness weight, and the resulting product is used as the video completeness score; Divide the course rating data by the full score data of the course rating, and then multiply the ratio by the course rating weight, and the resulting product is used as the course rating data score; Add the browsing duration score, video completeness score, and course rating data score and then multiply by the reciprocal of the number of knowledge points involved in the course to obtain the apportioned score of the user for this knowledge point in the course; Recalculate the apportioned scores for all courses containing this knowledge point, and sum the apportioned scores of each course to obtain the knowledge point mastery evaluation value. Among them, by using the course browsing duration, video viewing completeness, and course rating data collected from historical user learning data as independent variables, and the stage assessment score as the dependent variable, the least squares regression algorithm is used for parameter fitting, and the browsing duration weight, video completeness weight, and course rating weight are obtained according to the explanatory strength of each behavioral characteristic on the ability mastery effect, where the value ranges of the browsing duration weight, video completeness weight, and course rating weight are all [0, 1].

[0031] Among them, the specific formula for the knowledge point mastery evaluation value is: ; In the formula, represents the knowledge point mastery evaluation value, represents the course containing this knowledge point, represents the set of courses containing this knowledge point, represents the browsing duration weight, represents the course browsing duration, represents the historical average browsing duration of the course, represents the video completeness weight, represents the video viewing completeness, Indicates the course grading weight, Indicates the course grading data, Indicates the full score data of the course grading, Indicates the number of knowledge points involved in the course.

[0032] In this embodiment, Table 1 is the data table of the evaluation value of knowledge point mastery, which details the course set, the full score data of the course grading, the weight of the course browsing duration, the weight of the video viewing completeness, the course grading weight, and the finally calculated evaluation value of knowledge point mastery during the evaluation process for different knowledge points, and is used to quantify the learning and mastery of various military knowledge points by users. Among them: The course set corresponding to knowledge point C1 is B1, the full score data of the course grading is 10, the weight of the course browsing duration is 0.3, the weight of the video viewing completeness is 0.4, the course grading weight is 0.3, and the evaluation value of knowledge point mastery is 0.72; The course set corresponding to knowledge point C2 is B2, the full score data of the course grading is 10, the weight of the course browsing duration is 0.3, the weight of the video viewing completeness is 0.4, the course grading weight is 0.3, and the evaluation value of knowledge point mastery is 0.65; The course set corresponding to knowledge point C3 is B3, the full score data of the course grading is 10, the weight of the course browsing duration is 0.3, the weight of the video viewing completeness is 0.4, the course grading weight is 0.3, and the evaluation value of knowledge point mastery is 0.78; The course set corresponding to knowledge point C4 is B4, the full score data of the course grading is 10, the weight of the course browsing duration is 0.3, the weight of the video viewing completeness is 0.4, the course grading weight is 0.3, and the evaluation value of knowledge point mastery is 0.59; The course set corresponding to knowledge point C5 is B5, the full score data of the course grading is 10, the weight of the course browsing duration is 0.3, the weight of the video viewing completeness is 0.4, the course grading weight is 0.3, and the evaluation value of knowledge point mastery is 0.84.

[0033] Table 1 Data Table of the Evaluation Value of Knowledge Point Mastery [[ID=I7]]

[0034] As Figure 3 shown is the visualization bar chart of the evaluation value of knowledge point mastery. Combining with Table 2, it can be seen that there are obvious differences in the mastery of different knowledge points by users. Among them, the highest evaluation value of knowledge point mastery for knowledge point C2 is 0.84, indicating that the learning effect of users in the course set corresponding to this knowledge point is the best; The lowest mastery evaluation value of knowledge point C3 is only 0.59, indicating that the course learning behavior under this type of knowledge point is relatively weak; The mastery evaluation values of knowledge point C4 and knowledge point C1 are relatively high, and the learning state is stable; The evaluation value of knowledge point C5 is 0.65, which is at a relatively low level. Generally speaking, the visualization bar chart of the evaluation value of knowledge point mastery intuitively reflects the differences in the mastery of different knowledge points, and can be used as a basis for supplementary course recommendation and ability feedback adjustment.

[0035] In this implementation plan, by constructing the behavioral data relationship between users and knowledge points, based on the course browsing duration, video viewing completeness, and course rating data in the preprocessed user learning data, the historical average browsing duration of the course, the full score data of the course rating, and the number of knowledge points involved in the course are comprehensively introduced to form a multi-dimensional rating system based on the browsing duration score, video completeness score, and course rating data score, realizing the refined modeling and quantitative evaluation of the user's learning and mastery of each knowledge point, effectively enhancing the discriminant ability and dynamic expression ability of the knowledge point mastery evaluation value, and laying a highly reliable evaluation foundation for subsequent accurate recommendation and feedback regulation.

[0036] Specifically, the specific steps for constructing the time series knowledge point rating set are as follows: After calculating the knowledge point mastery evaluation value, the knowledge point mastery evaluation value is compared with the mastery threshold in real time: When the knowledge point mastery evaluation value is greater than or equal to the mastery threshold, it is marked as a normally mastered knowledge point and no adjustment is required; when the knowledge point mastery evaluation value is less than the mastery threshold, it is marked as an abnormally mastered knowledge point, a prompt is given for the user's current learning path, and a recommendation instruction to add supplementary courses related to this knowledge point is generated; The knowledge point mastery evaluation value after comparison is processed in time series: The knowledge point mastery evaluation value generated by the user in each learning behavior is recorded in chronological order, and the corresponding timestamp, course code, and knowledge point label are marked; All the knowledge point mastery evaluation values of the same user on the same knowledge point are sorted in chronological order to construct the knowledge point mastery evaluation value sequence of the user on this knowledge point; The knowledge point mastery evaluation value sequences of the user on all knowledge points are sorted and organized to form the time series knowledge point mastery evaluation value set of the user under the entire knowledge structure.

[0037] In this implementation plan, by comparing the knowledge point mastery evaluation value with the mastery threshold in real time, abnormally mastered knowledge points are identified in a timely manner, and combined with the learning path, a supplementary course recommendation instruction is dynamically generated, realizing the active intervention of learning blind spots; at the same time, by marking and sorting the knowledge point mastery evaluation value in chronological order, a time series knowledge point mastery evaluation value set covering all knowledge points is constructed, effectively enhancing the tracking and expression ability of the user's knowledge mastery evolution process, and providing high-resolution, time-series data support for subsequent stage ability evaluation and personalized recommendation strategies.

[0038] Specifically, based on the time series knowledge point scoring set, the abnormal mastery of knowledge points in the user's current stage is identified, and differentiated analysis is carried out in combination with the behavioral data relationship. The specific steps for evaluating the recommendation value of candidate courses are as follows: Combine the time series knowledge point mastery evaluation value set of the user under the entire knowledge structure, extract the set of knowledge points marked as abnormal mastery in the current period, and count the number of knowledge points marked as abnormal mastery; group abnormal records according to knowledge point labels, identify the knowledge point labels that are continuously marked as abnormal within a short period of time within the same user, and form a set of key supplementary knowledge points for the user's current stage; based on the identified set of key supplementary knowledge points, review all courses associated with each knowledge point, eliminate courses that the user has completed and courses that he has not watched completely, and include the remaining courses in the candidate course set; for For each course in the candidate course set, the average historical course rating is calculated, the number of overlaps between the course-related knowledge points and the supplementary knowledge point set is extracted, and the course is analyzed to see if it is worth recommending; the square value of the number of overlaps between the course-related knowledge points and the supplementary knowledge point set is calculated, the product of the number of knowledge points involved in the course and the number of knowledge points marked as abnormally mastered is calculated, this square value is divided by this product, and the ratio is used as the knowledge point matching score; the course browsing time is calculated and divided by the average historical course browsing time, and this ratio is added to the video viewing completeness to obtain the learning input score; the course rating data plus the course historical average rating is calculated, and the sum is divided by twice the full score of the course rating data to obtain the course rating data performance score; the knowledge point matching score, learning input score, and course rating data performance score are multiplied together to obtain the course recommendation strength assessment value.

[0039] The specific calculation formula for the course recommendation strength evaluation value is: ; Where, Indicates the evaluation value of the course recommendation strength, Indicates the number of overlaps between the course-related knowledge points and the set of supplementary knowledge points. Indicates the number of knowledge points involved in the course, Indicates the number of knowledge points marked as abnormally mastered, Indicates the course browsing time. Indicates the average browsing time of the course history. Indicates the video viewing completeness. Represents course rating data, Indicates the historical average score of the course. Indicates the full score data of the course.

[0040] In this implementation plan, by constructing a time series knowledge point mastery evaluation value set, accurately identifying abnormal knowledge points in the current stage, and conducting fine-grained differential analysis in combination with user behavior data relationships, a key supplementary knowledge point set is formed; on this basis, the number of overlapping knowledge points between the course-related knowledge points and the supplementary knowledge point set, the course browsing duration, the video viewing completeness, and the course rating data are introduced, and a comprehensive index including the knowledge point matching score, the learning input score, and the course rating data performance score is constructed to quantify the course recommendation intensity evaluation value of the candidate courses, so as to enhance the pertinence and effectiveness of the recommended course screening.

[0041] Specifically, the specific steps for partitioning the candidate courses and generating recommended courses are as follows: Based on the course recommendation intensity evaluation value, construct a course recommendation intensity evaluation index system by analyzing the matching degree between the course-related knowledge points and the supplementary knowledge point set, the course browsing behavior input, and the course rating data performance, and construct a multi-dimensional course recommendation popularity partition according to the course recommendation threshold; among them, the course recommendation threshold includes a primary course recommendation threshold and a secondary course recommendation threshold, which are dynamically adjusted according to the course feedback records in the historical recommendation results and the fluctuation trend of the knowledge point mastery evaluation value; when the course recommendation intensity evaluation value is greater than or equal to the secondary course recommendation threshold, it is determined as a high-popularity course, directly included in the current recommended courses, and given priority in sorting and pushing, and at the same time trigger the subsequent monitoring process of the mastery feedback response value in the recommendation strategy module; when the course recommendation intensity evaluation value is greater than the primary course recommendation threshold and less than the secondary course recommendation threshold, it is determined as a medium-popularity course, enter the waiting area, conduct theme aggregation through the supplementary knowledge point labels covered by the course, and give priority to matching with the knowledge point labels not covered in the user's learning record, and recommend 1 to 2 courses according to the label distribution to expand the recommendation coverage and maintain the recommendation diversity; when the course recommendation intensity evaluation value is less than or equal to the primary course recommendation threshold, it is determined as a low-popularity course, not included in the current recommended courses, but recorded as a backfill candidate resource for subsequent mastery feedback anomalies; generate the recommended courses for the user at the current stage according to the multi-dimensional course recommendation popularity partition result, and send the recommended courses to the user learning interface; mark the supplementary knowledge point labels covered by each course in the recommended course list, and identify whether it is a course not learned in the user's historical learning path to assist the user in course selection and path judgment.

[0042] In this implementation plan, by introducing the course recommendation intensity evaluation value and constructing the course recommendation threshold interval, multi-dimensional heat zone management of candidate courses is realized. It not only fully considers the matching degree between the associated knowledge points of the course and the set of supplementary knowledge points, the input of course browsing behavior and the performance of course scoring data, but also combines the knowledge point tags not covered in the user's learning record, enhancing the adaptability and expansibility of the recommended content. At the same time, different heat zone courses are given differential push mechanisms, which are linked with the subsequent monitoring process of the mastery feedback response value, constructing a dynamic adjustment and feedback closed-loop mechanism, providing data support and strategy guarantee for the accuracy, controllability and dynamic sustainable optimization of the recommendation results.

[0043] Specifically, the specific steps to quantify the response difference between the recommended course and the mastery effect from the proportion dimension are as follows: After the user completes the learning of the recommended course, extract the course code, course browsing duration, and supplementary knowledge point tags covered in the learning of this recommended course, and establish the content structure of the recommended course learning; Synchronously record the course test score data after the recommended course, and update the corresponding knowledge point mastery evaluation value in the knowledge point dimension; Construct the comparison relationship of the knowledge point mastery evaluation value before and after the recommendation according to the user and the knowledge point; According to the comparison relationship of the knowledge point mastery evaluation value before and after the recommendation, quantify the response difference between the recommended course and the mastery effect: Subtract the knowledge point mastery evaluation value of this knowledge point before the recommended course is issued from the knowledge point mastery evaluation value of this knowledge point after the recommended course is issued, then divide by the full score data of the course score, and then subtract the ratio of the course test score data to the full score data of the course test from this ratio. The absolute value of the difference is used as the mastery change ratio; Add 1 to the number of recommended courses involving this knowledge point in the recommended course, then take the logarithm to the base 2, and add 1 to the result as the course coverage factor; Multiply the mastery change ratio and the course coverage factor to obtain the mastery feedback response value.

[0044] Among them, the specific calculation formula of the mastery feedback response value is: ; In the formula, represents the mastery feedback response value, represents the knowledge point mastery evaluation value after the recommendation, represents the knowledge point mastery evaluation value before the recommendation, represents the full score data of the course score, represents the course test score data, represents the full score data of the course test, represents the number of recommended courses involving this knowledge point in the recommended course.

[0045] In this embodiment, Table 2 is the data table of the mastery feedback response value, which records the evaluation data of 5 knowledge points. Each knowledge point corresponds to a series of calculation results, including the mastery evaluation values of the knowledge points before and after the recommended courses and the weight values of various items related to the courses. The specific data is as follows: Knowledge point A: The mastery evaluation value of the knowledge point before recommendation is 0.42, the mastery evaluation value of the knowledge point after recommendation is 0.68, the full score data of the course score is 5.00, the score data of the course test is 4.20, the full score data of the course test is 5.00, the number of recommended courses is 2, and the finally calculated mastery feedback response value is 2.04. Knowledge point B: The mastery evaluation value of the knowledge point before recommendation is 0.51, the mastery evaluation value of the knowledge point after recommendation is 0.55, the full score data of the course score is 5.00, the score data of the course test is 4.00, the full score data of the course test is 5.00, the number of recommended courses is 1, and the finally calculated mastery feedback response value is 1.58. Knowledge point C: The mastery evaluation value of the knowledge point before recommendation is 0.63, the mastery evaluation value of the knowledge point after recommendation is 0.76, the full score data of the course score is 5.00, the score data of the course test is 4.40, the full score data of the course test is 5.00, the number of recommended courses is 3, and the finally calculated mastery feedback response value is 2.68. Knowledge point D: The mastery evaluation value of the knowledge point before recommendation is 0.58, the mastery evaluation value of the knowledge point after recommendation is 0.60, the full score data of the course score is 5.00, the score data of the course test is 4.30, the full score data of the course test is 5.00, the number of recommended courses is 2, and the finally calculated mastery feedback response value is 1.95. Knowledge point E: The mastery evaluation value of the knowledge point before recommendation is 0.49, the mastery evaluation value of the knowledge point after recommendation is 0.57, the full score data of the course score is 5.00, the score data of the course test is 4.10, the full score data of the course test is 5.00, the number of recommended courses is 1, and the finally calculated mastery feedback response value is 1.61.

[0046] Table 2 Data Table of Mastery Feedback Response Value

[0047] As Figure 4 shown, it is the line chart of the mastery evaluation value and the mastery feedback response value before and after the recommended courses. Combining Table 2 and Figure 4It can be seen that the changing trends of the knowledge point mastery evaluation values and the corresponding mastery feedback response values before and after the recommended courses are learned for the five knowledge points. In knowledge point C, the change in the knowledge point mastery evaluation value before and after the recommended course is relatively large, and the highest mastery feedback response value is 2.68, indicating that the recommended course has caused an obvious mastery deviation. In contrast, the change range of the knowledge point mastery evaluation value of knowledge point B before and after the recommendation is the smallest, and the mastery feedback response value is also relatively low at 1.58, indicating that the recommendation effect is relatively stable. In addition, knowledge points A and D also show medium-level response values, while knowledge point E shows relatively small changes and low mastery feedback response values. Overall, the data in the figure clearly reflect the differences in the mastery change range and feedback intensity of different knowledge points under the intervention of the recommended content.

[0048] In this implementation plan, by constructing the control relationship of the knowledge point mastery evaluation values before and after the recommendation, quantifying the response difference between the recommended course and the mastery effect, the diagnostic analysis ability of the recommended content is improved. Introducing the mastery change ratio and the course coverage factor can reflect the actual improvement of the knowledge point mastery degree and the impact of the course distribution on the learning effect from the proportional dimension, ensuring that the generation of the mastery feedback response value is more accurate and discriminative, providing a quantifiable basis for the feedback correction of the subsequent recommended content and the adjustment of the recommendation strategy, and enhancing the system's tracking ability and intervention rationality for the user's learning process.

[0049] Specifically, the specific steps to update the recommended tags and user records and improve the recommended data loop are as follows: After calculating and obtaining the mastery feedback response value, the mastery feedback response value is compared with the mastery feedback threshold in real time: among them, the mastery feedback threshold includes the first-level mastery feedback threshold and the second-level mastery feedback threshold; when the mastery feedback response value is greater than or equal to the second-level mastery feedback threshold, the feedback is normal, and the knowledge points covered in the current course are marked as recommended valid tags, enhancing the recommended priority of the current course in the candidate course ranking for subsequent similar users; when the mastery feedback response value is greater than the first-level mastery feedback threshold and less than the second-level mastery feedback threshold, the feedback is lagging, and the current course is marked as an intermediate transition course and incorporated into the next recommendation cycle as a continuous observation object for delayed supplementary content; when the mastery feedback response value is less than or equal to the first-level mastery feedback threshold, the feedback is abnormal, and the knowledge points covered in the current course are marked as recommended offset tags, marking the current course as a recommended insensitive course, and preferentially avoiding courses with similar teaching structures in subsequent supplementary course screening; According to the user dimension, the mastery feedback response values generated by the same knowledge point in different recommendation cycles are continuously arranged to construct a mastery feedback response value sequence; for knowledge points with the mastery feedback response value less than the first-level mastery feedback threshold three times or more, extract the course code and course score data corresponding to the knowledge point, and recalculate the knowledge point mastery evaluation value of the user in this course based on the original course browsing duration, video viewing completeness, and course score data, replacing the original result for feedback correction; write the course code, knowledge point tags, course browsing duration, course score data, video viewing completeness, mastery feedback response value, course test score data, and evaluation value change results generated by each round of recommended course learning into the user recommendation record; in the process of generating a new round of recommendations, preferentially read the user recommendation record, dynamically adjust the initial recommended course ranking, popularity differentiation strategy, and recommended threshold calculation logic, update the binding strength between the candidate course and the key knowledge point, and complete the active adaptation adjustment of the recommended path, thereby realizing the recommendation self-closed-loop mechanism based on ability feedback.

[0050] In this implementation, by introducing a multi-level comparison mechanism that grasps the feedback response value and the feedback threshold, refined feedback classification and label update of the course recommendation results are achieved. It can dynamically identify the effectiveness, lag, and deviation of recommendations, and clearly mark the course recommendation status with recommended effective labels, intermediate transition courses, and recommended deviation labels. At the same time, by combining the construction of the feedback sequence on the user dimension and the monitoring of stage fluctuations, the system improves the recognition depth of the changes in knowledge point mastery and promotes the iterative correction of the original knowledge point mastery evaluation value. By continuously writing course codes, knowledge point labels, course browsing duration, course score data, video viewing completeness, feedback response value for mastery, course test score data, and evaluation value change results, the system effectively accumulates recommendation record data and dynamically updates the course sorting strategy and recommendation threshold in the next round of recommendation generation, strengthening the adaptability between candidate courses and key knowledge points, and finally realizing the adaptive adjustment of the recommendation path driven by the feedback response value for mastery and the optimization of the data closed-loop.

[0051] As Figure 2 shown, the second aspect of the present invention provides a personalized course recommendation system based on a knowledge graph, including: a user learning data collection and preprocessing module, a knowledge point mastery evaluation module, a supplementary course recommendation generation module, and a recommendation effect feedback correction module, where: The user learning data collection and preprocessing module is used to obtain user learning data by collecting real-time front-end buried point and background access log data of the learning platform, and preprocess the user learning data to obtain preprocessed user learning data; The knowledge point mastery evaluation module is used to construct the behavioral data relationship between users and knowledge points, perform normalized splitting on the preprocessed user learning data, evaluate the user's mastery of knowledge points, and construct a time series knowledge point score set; The supplementary course recommendation generation module is used to identify abnormal knowledge points mastered by the user at the current stage based on the time series knowledge point score set, conduct differential analysis in combination with the behavioral data relationship, evaluate the recommendation value of candidate courses, partition the candidate courses, and generate recommended courses; The recommendation effect feedback correction module is used to quantify the response difference between the recommended courses and the mastery results from the ratio dimension, update the recommendation labels and user records, and improve the recommendation data cycle.

[0052] In this implementation, by systematically integrating the user learning data collection and preprocessing module, the knowledge point mastery evaluation module, the supplementary course recommendation generation module, and the recommendation effect feedback and correction module, the structured cleaning of user learning data, the refined evaluation of knowledge point mastery, the differential recommendation of supplementary courses, and the dynamic feedback and correction of recommendation effects are realized. The system can continuously track the time series changes in the knowledge point mastery status, conduct the recommendation value analysis and heat zone division of candidate courses in combination with behavior data, and realize the intelligent generation of course recommendations. At the same time, a mastery feedback response mechanism is introduced to evaluate the recommendation effectiveness from the proportion dimension and update the recommendation labels and user records, forming a closed-loop process of data collection, analysis, recommendation, feedback, and re-optimization, significantly improving the matching degree between the recommended content and the actual mastery needs of users.

[0053] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0054] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and changes can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A personalized course recommendation method based on a knowledge graph, characterized in that, It includes the following steps: S1. By collecting the front-end buried points and background access log data of the learning platform in real time, obtain the user learning data, and preprocess the user learning data to obtain the preprocessed user learning data; S2. Build the behavioral data relationship between users and knowledge points, perform normalized splitting on the preprocessed user learning data, evaluate the user's mastery of knowledge points, and build a time series knowledge point scoring set; S3. Based on the time series knowledge point scoring set, identify the abnormally mastered knowledge points of the user at the current stage, conduct differential analysis in combination with the behavioral data relationship, evaluate the recommendation value of the candidate courses, partition the candidate courses, and generate recommended courses; S4. Quantify the response difference between the recommended courses and the mastery effectiveness from the proportion dimension, update the recommendation tags and user records, and improve the recommendation data loop.

2. The personalized curriculum recommendation method based on a knowledge graph according to claim 1, wherein: The specific steps for obtaining the user learning data by collecting the front-end buried points and background access log data of the learning platform in real time are as follows: By collecting the front-end buried points and background access log data of the learning platform in real time, obtain the user learning data. The user learning data includes course code, course browsing duration, course access times, historical average course browsing duration, video viewing completeness, course rating data, course rating full score data, course test score data, course test full score data, and the number of knowledge points involved in the course. Among them, the course code is extracted from the course name field in the course access log; the course browsing duration is obtained by recording the difference between the start time and the end time when the user enters the course page each time and summing up by course; the course access times are obtained by counting the total number of times the user accesses a certain course page; the historical average course browsing duration is obtained by dividing the total browsing duration of the user for the same course within a week by the course access times; the video viewing completeness is obtained by calculating the ratio of the actual playing duration recorded by the video player to the total duration of the course video; the course rating data and course rating full score data are obtained by extracting the rating field submitted by the user after completing the course learning; the course test score data and course test full score data are obtained by recording the scoring results of the user participating in the in-course quiz after completing the course; the number of knowledge points involved in the course is obtained by analyzing the mapping relationship between the course and the knowledge points in the course resource management database and counting the number of knowledge point labels covered by each course.

3. The personalized course recommendation method based on a knowledge graph according to claim 1, characterized in that: The specific steps for preprocessing the user learning data to obtain the preprocessed user learning data are as follows: Combining the median absolute deviation method of sliding time window with the isolation forest algorithm, local and global anomaly detection is carried out on user learning data, and abnormal user learning data including abnormal long-term hanging up and skipping classes to brush progress is removed; through the similar user collaborative completion strategy and the exponentially weighted moving average algorithm, data filling is performed on the missing field problems of user learning data caused by system delay, missing data points or terminal fluctuations; fitting and denoising of user learning data are carried out through the Bayesian dynamic smoothing method and the locally weighted regression filtering algorithm; transformation and range compression of user learning data are carried out through exponential transformation combined with the maximum-minimum normalization method to achieve normalization processing.

4. The personalized curriculum recommendation method based on a knowledge graph according to claim 1, characterized in that: The specific steps for constructing the behavioral data relationship between users and knowledge points, normalizing and splitting the preprocessed user learning data, and evaluating the user's mastery of knowledge points are as follows: According to the knowledge point labels marked in each course, the corresponding knowledge point set is extracted; combined with the course codes in the user learning record, the knowledge point labels included in the courses learned by the user are searched one by one and merged at the user dimension to obtain the complete knowledge point set learned by the user; Taking the user name and the knowledge point label as corresponding items, the association records between the user and the knowledge point are sorted out to form the behavioral correspondence between the user and the knowledge point during the learning process; Based on the established behavioral correspondence between the user and the knowledge point, for each knowledge point, by integrating the preprocessed user learning data such as the course browsing duration, video viewing completeness and course rating data generated by the user in the courses containing the same knowledge point, the user's mastery of this knowledge point is evaluated: for a course containing this knowledge point, calculate the ratio between the course browsing duration and the historical average browsing duration of the course, and multiply this ratio by the browsing duration weight, and the obtained product is used as the browsing duration score; multiply the video viewing completeness by the video completeness weight, and the obtained product is used as the video completeness score; divide the course rating data by the full score data of the course rating, and then multiply the ratio by the course rating weight, and the obtained product is used as the course rating data score; add the browsing duration score, the video completeness score and the course rating data score and then multiply by the reciprocal of the number of knowledge points involved in the course to obtain the apportioned score of the user for this knowledge point in the course; Recalculate the apportioned scores for all courses containing this knowledge point, and sum up the apportioned scores of each course to obtain the knowledge point mastery evaluation value.

5. The personalized course recommendation method based on a knowledge graph according to claim 1, wherein: The specific steps for constructing the time series knowledge point score set are as follows: After calculating the knowledge point mastery evaluation value, compare the knowledge point mastery evaluation value with the mastery threshold in real time: when the knowledge point mastery evaluation value is greater than or equal to the mastery threshold, it is marked as a normally mastered knowledge point and no adjustment is required; when the knowledge point mastery evaluation value is less than the mastery threshold, it is marked as an abnormally mastered knowledge point, a prompt is given to the user's current learning path, and a recommendation instruction to add supplementary courses related to this knowledge point is generated; The compared knowledge point mastery evaluation values are processed in time series: the knowledge point mastery evaluation values generated by the user in each course learning behavior are recorded in chronological order, and the corresponding timestamps, course codes and knowledge point tags are marked; all knowledge point mastery evaluation values of the same user on the same knowledge point are sorted in chronological order to construct the user's knowledge point mastery evaluation value sequence on this knowledge point; the user's knowledge point mastery evaluation value sequence on all knowledge points is aggregated and organized to form a time series knowledge point mastery evaluation value set of the user in the entire knowledge structure.

6. The personalized curriculum recommendation method based on a knowledge graph according to claim 1, wherein: The specific steps of identifying the abnormal knowledge points mastered by the user at the current stage based on the time series knowledge point scoring set, conducting differentiation analysis based on the behavioral data relationship, and evaluating the recommendation value of the candidate courses are as follows: Combine the user's time series knowledge point mastery evaluation value set under the entire knowledge structure, extract the set of knowledge points marked as abnormally mastered in the current period, and count the number of knowledge points marked as abnormally mastered; Group abnormal records by knowledge point tags, identify knowledge point tags that are continuously marked as abnormal within a short period of time within the same user, and form a set of key supplementary knowledge points for the user at the current stage; Based on the identified set of key supplementary knowledge points, we review all courses associated with each knowledge point, remove courses that the user has completed or has not fully watched, and include the remaining courses in the candidate course set. For each course in the candidate course set, we calculate the course's historical average rating, extract the number of overlaps between the course's associated knowledge points and the set of supplementary knowledge points, and analyze whether the course is worth recommending. Calculate the square of the number of overlaps between the course-related knowledge points and the set of supplementary knowledge points. Calculate the product of the number of knowledge points involved in the course and the number of knowledge points marked as abnormally mastered. Divide this square by this product, and use the ratio as the knowledge point matching score. Calculate the course browsing time divided by the average browsing time of the course history, and add this ratio to the video viewing completeness to get the learning investment score; Calculate the course rating data plus the historical average course rating, and divide the sum by twice the full score of the course rating data as the course rating data performance score; multiply the knowledge point matching score, learning input score and course rating data performance score to obtain the course recommendation strength assessment value.

7. A personalized course recommendation method based on a knowledge graph according to claim 1, characterized in that: The specific steps of partitioning candidate courses and generating recommended courses are as follows: Based on the evaluation value of course recommendation intensity, a multi-dimensional course recommendation popularity partition is constructed according to the course recommendation threshold: among them, the course recommendation threshold includes a primary course recommendation threshold and a secondary course recommendation threshold; when the evaluation value of course recommendation intensity is greater than or equal to the secondary course recommendation threshold, it is determined as a high-popularity course, which is directly included in the current recommended courses and given priority in the push order; when the evaluation value of course recommendation intensity is greater than the primary course recommendation threshold and less than the secondary course recommendation threshold, it is determined as a medium-popularity course, which enters the waiting area, is subject to topic aggregation through the supplementary knowledge point tags covered by the course, and is preferentially matched according to the user's historical unlearned content, and 1 to 2 courses are recommended according to the tag distribution to expand the recommendation coverage; when the evaluation value of course recommendation intensity is less than or equal to the primary course recommendation threshold, it is determined as a low-popularity course and does not enter the current recommended courses; Generate the recommended courses for the user at the current stage according to the results of the multi-dimensional course recommendation popularity partition, and send the recommended courses to the user's learning interface; mark the supplementary knowledge point tags covered by each course in the recommended course list.

8. The personalized course recommendation method based on a knowledge graph according to claim 1, wherein: The specific steps for quantifying the response difference between the recommended courses and the mastery effectiveness from the proportion dimension are as follows: After the user completes the study of the recommended courses, extract the course codes, course browsing durations, and supplementary knowledge point tags covered by the current recommended course study to establish the content structure of the recommended course study; synchronously record the course test score data after the recommended courses, and update the corresponding knowledge point mastery evaluation value in the knowledge point dimension; construct the comparison relationship of the knowledge point mastery evaluation value before and after the recommendation according to the user and the knowledge points; According to the comparison relationship of the knowledge point mastery evaluation value before and after the recommendation, quantify the response difference between the recommended courses and the mastery effectiveness: subtract the knowledge point mastery evaluation value of the knowledge point after the recommended course is issued from the knowledge point mastery evaluation value of this knowledge point before the recommended course is issued, then divide by the full score data of the course score, and then subtract the ratio of the course test score data to the full score data of the course test from this ratio, and take the absolute value of the difference as the mastery change ratio; add 1 to the number of recommended courses involving this knowledge point in the recommended courses, take the logarithm to the base 2, and add 1 to the result as the course coverage factor; multiply the mastery change ratio and the course coverage factor to obtain the mastery feedback response value.

9. The personalized course recommendation method based on a knowledge graph according to claim 1, wherein: The specific steps for updating the recommended tags and user records and improving the recommended data loop are as follows: After calculating and obtaining the mastery feedback response value, the mastery feedback response value is compared with the mastery feedback threshold in real time: among them, the mastery feedback threshold includes a primary mastery feedback threshold and a secondary mastery feedback threshold; when the mastery feedback response value is greater than or equal to the secondary mastery feedback threshold, the feedback is normal, and the knowledge points covered in the current course are marked with a recommended valid label, improving the recommended priority of the current course in the candidate course ranking for subsequent similar users; when the mastery feedback response value is greater than the primary mastery feedback threshold and less than the secondary mastery feedback threshold, the feedback is lagged, and the current course is marked as an intermediate transition course and included in the next recommendation cycle as a continuous observation object for delayed supplementary content; when the mastery feedback response value is less than or equal to the primary mastery feedback threshold, the feedback is abnormal, and the knowledge points covered in the current course are marked with a recommended deviation label, marking the current course as a recommended insensitive course, and preferentially avoiding courses with similar teaching structures in the subsequent supplementary course screening. According to the user dimension, the mastery feedback response values generated by the same knowledge point in different recommendation cycles are continuously arranged to construct a mastery feedback response value sequence; for knowledge points with the mastery feedback response value less than the primary mastery feedback threshold three or more times, the course code and course score data corresponding to the knowledge points are extracted, and the knowledge point mastery evaluation value of the user on this course is recalculated based on the original course browsing duration, video viewing completeness, and course score data, replacing the original result for feedback correction. The course code, knowledge point label, course browsing duration, course score data, video viewing completeness, mastery feedback response value, course test score data, and evaluation value change results generated by each round of recommended course learning are written into the user recommendation record; in the process of generating a new round of recommendations, the user recommendation record is preferentially read, the initial recommended course sorting, popularity differentiation strategy, and recommended threshold calculation logic are dynamically adjusted, the binding strength between the candidate course and the key knowledge point is updated, and the active adaptation adjustment of the recommendation path is completed, so as to realize the recommendation self-closed-loop mechanism based on ability feedback.

10. A personalized course recommendation system based on a knowledge graph, characterized in that: Including: User learning data collection and preprocessing module, knowledge point mastery evaluation module, supplementary course recommendation generation module, and recommendation effect feedback correction module, where: The user learning data collection and preprocessing module is used to obtain user learning data by real-time collecting the front-end buried point and background access log data of the learning platform, and preprocess the user learning data to obtain the preprocessed user learning data. The knowledge point mastery evaluation module is used to construct the behavioral data relationship between the user and the knowledge point, evaluate the user's mastery of the knowledge point based on the preprocessed user learning data, and construct a time series knowledge point score set. The supplementary course recommendation generation module is used to identify the abnormal mastery knowledge points of the user at the current stage based on the time series knowledge point score set, analyze whether the candidate course is worth recommending, partition the candidate course, and generate a recommended course. The recommendation effect feedback correction module is used to quantify the response difference between the recommended course and the mastery effect, update the recommendation label and user record, and improve the recommendation data loop.

Images