Learning resource intelligent recommendation system based on data analysis

Abstract

The invention relates to the technical field of learning resource recommendation, and particularly discloses a learning resource intelligent recommendation system based on data analysis, and the system comprises a behavior collection module which is used for collecting user video learning data, exercise interaction data and learning rhythm data in real time; the real-time analysis module comprises a knowledge point mastering evaluation unit and a learning load evaluation unit, the knowledge point mastering evaluation unit is used for evaluating the mastering degree of the user on each knowledge point, and the learning load evaluation unit is used for evaluating the current learning pressure of the user; multi-aspect learning data are widely collected through the behavior collection module, the knowledge point mastering degree and learning pressure are accurately evaluated through the real-time analysis module, the exercise difficulty and the follow-up learning sequence are flexibly adjusted through the resource recommendation module according to the analysis result, learning resource recommendation is in personalized adaptation with a user, and the learning efficiency is improved. Appropriate knowledge points and exercises are accurately recommended to the user, and the effect of remarkably improving learning efficiency and pertinence is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of learning resource recommendation, and specifically to an intelligent learning resource recommendation system based on data analysis. Background Art

[0002] With the rapid development of information technology and the acceleration of the global digital transformation of education, online education has shown an explosive growth trend. The popularization of 5G networks, the wide application of intelligent terminal devices, and the maturity of cloud computing technology provide strong technical support for the online transmission of educational resources, enabling a new learning mode based on video online courses and online exercises to break through the time and space limitations of traditional classrooms and become the mainstream learning form covering all fields such as education, vocational training, and higher education. Traditional online education systems mostly adopt a preset linear learning path, and knowledge points and exercises are pushed in a fixed order, resulting in a low matching degree between the difficulty of the recommended knowledge points or exercises and the actual mastery level of users. It may cause users to repeatedly learn simple content they have already mastered, generating a sense of boredom, or users are forced to challenge high-difficulty content beyond their current capabilities, reducing their learning enthusiasm due to difficulty in understanding. This not only ignores the differences in individual knowledge mastery but also easily leads to users copying answers from each other through standardized paths, weakening the learning effect. In addition, traditional online education systems lack dynamic assessment of users' learning loads and do not consider the learning pressure behind behaviors such as spending too long answering questions, frequently modifying answers, or unexpectedly switching knowledge points, resulting in the recommendation strategy failing to adapt to the real-time state of learners and affecting learning efficiency and experience. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent learning resource recommendation system based on data analysis to solve the following technical problems: How to solve the problem of insufficient personalized adaptation in the learning resource recommendation of traditional online education and the inability to accurately recommend appropriate knowledge points and exercises to users.

[0004] The purpose of the present invention can be achieved through the following technical solutions: An intelligent learning resource recommendation system based on data analysis, the system includes: A behavior collection module for real-time collecting user video learning data, exercise interaction data, and learning rhythm data; A real-time analysis module, including a knowledge point mastery assessment unit and a learning load assessment unit. The knowledge point mastery assessment unit is used to assess the mastery degree of users for each knowledge point, and the learning load assessment unit is used to assess the current learning pressure of users; The resource recommendation module adjusts the difficulty of the recommended exercises according to the data obtained from the analysis of the knowledge point mastery evaluation unit; and adjusts the knowledge points and exercise order for subsequent learning according to the data obtained from the analysis of the knowledge point mastery evaluation unit and the learning load evaluation unit.

[0005] Further, the process of evaluating the user's mastery of each knowledge point includes:

[0006]

[0007]

[0008] The mastery evaluation coefficient of each knowledge point is obtained through analysis and calculation by formulas (1)-(3) ; Among them, , are the first weight coefficients, is the number of playback times during the learning process of the learning video corresponding to the knowledge point, is the playback decay coefficient, is the duration influence factor of the learning video corresponding to the knowledge point, is the answering correct rate of the exercises corresponding to the knowledge point, is the number of modifications during the answering process of the exercises corresponding to the knowledge point, is the modification times influence factor, is the actual learning duration of the learning video corresponding to the knowledge point, is the reference learning duration of the learning video corresponding to the knowledge point, is the number of pauses during the learning process of the learning video corresponding to the knowledge point, is the pause influence factor.

[0009] Further, the process of evaluating the user's current learning pressure includes:

[0010] The user's current learning pressure load index is obtained through analysis and calculation by formula (4) ; Among them, is the actual time used by the user to complete the previous exercise, is the reference answering time corresponding to the previous exercise, is the number of times the user modifies the answer during the answering process of a certain number of recent exercises, is the number of exercises that the user actively skips during the answering process of a certain number of recent exercises, is the skip exercise influence factor, is the preset reference number of times, The number of times of switching knowledge points by the user during the learning process that is not in accordance with the system-recommended path, is the total number of times of switching knowledge points by the user during the learning process, 、 、 is the second weight coefficient.

[0011] Further, the process of adjusting the difficulty of the recommended exercises includes:

[0012]

[0013] The recommended difficulty is obtained through analysis and calculation by formulas (5)-(6) ; wherein, is the initial benchmark difficulty corresponding to the knowledge point, is the difficulty adjustment coefficient, 、 、 are the difficulty adjustment slope coefficients, is the preset first knowledge point mastery evaluation coefficient, is the preset second knowledge point mastery evaluation coefficient.

[0014] Further, the process of adjusting the knowledge points and exercise order for subsequent learning also includes: When the user continuously completes a certain number of exercises of the same difficulty, the difficulty adjustment slope coefficient is corrected according to the user's answering performance;

[0015] The corrected difficulty adjustment slope coefficient is obtained through analysis and calculation by formula (7) ; wherein, is the correct rate of a certain number of completed exercises, is the preset basic correct rate, is the correction amplitude coefficient.

[0016] Further, the process of adjusting the knowledge points and exercise order for subsequent learning includes: Based on the knowledge graph topological structure, an initial learning path is generated according to the knowledge point dependency relationship; The current learning stress load index of the user is compared with the preset learning stress load index critical value ; If , then

[0017]

[0018]

[0019]

[0020] The adjusted learning path is obtained through analytical calculation by formulas (8)-(11). ; Among them, is the knowledge point coherence score of the learning path, is the learning novelty score, is the dynamic adjustment factor, is the average value of the mastery evaluation coefficients of each knowledge point in the learning path, is the preset critical value of the learning pressure load index, 、 are the third weight coefficients, is the mastery evaluation coefficient corresponding to the i-th knowledge point in the learning path, is the initial baseline difficulty corresponding to the i-th knowledge point, is the initial baseline difficulty corresponding to the (i - 1)-th knowledge point, is the proportion of knowledge points that the user has not learned in the learning path, is the degree of repetition between the learning path and the user's historical learning path, is the basic correction factor, is the learning path that maximizes the function output value; If , then only the knowledge points of are retained in the learning path, and the initial learning path length is reduced to half of the original length, is the preset basic mastery evaluation coefficient of the knowledge point.

[0021] Furthermore, the process of evaluating the user's mastery of each knowledge point further includes: Analyze the mastery evaluation coefficients of the knowledge points that the user has not learned; Extract all the prerequisite dependent knowledge points of the target knowledge point

[0022]

[0023] The mastery evaluation coefficient of the knowledge points that the user has not learned is obtained through analytical calculation by formula (12). ; Among them, is the mastery evaluation coefficient of the prerequisite dependent knowledge point , is the prerequisite dependent knowledge point Evaluation coefficient for mastering knowledge points not learned before Influence weight.

[0024] Advantages of the present invention: (1) By widely collecting multi-faceted learning data through the behavior acquisition module, accurately evaluating the mastery of knowledge points and learning pressure through the real-time analysis module, and flexibly adjusting the difficulty of exercises and subsequent learning order according to the analysis results through the resource recommendation module, the learning resource recommendation is personalized and adapted to the user, and appropriate knowledge points and exercises are accurately recommended to the user, achieving the effect of significantly improving learning efficiency and pertinence, and solving the problem of insufficient personalized adaptation in traditional online education learning resource recommendation and the inability to accurately recommend appropriate knowledge points and exercises to users. Brief description of the drawings

[0025] The present invention will be further described below with reference to the drawings.

[0026] Figure 1 is a schematic block diagram of an intelligent learning resource recommendation system based on data analysis proposed by the present invention. Specific implementation manners

[0027] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the 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 of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0028] Please refer to Figure 1 As shown, in one embodiment, an intelligent learning resource recommendation system based on data analysis is provided, and the system includes: A behavior acquisition module for real-time collecting user video learning data, exercise interaction data and learning rhythm data. The user video learning data includes, but is not limited to, viewing duration, pause times, and playback frequency. The exercise interaction data includes, but is not limited to, answering accuracy rate, answering speed, and modification times. The learning rhythm data includes, but is not limited to, chapter switching frequency and knowledge point jumping situation; A real-time analysis module including a knowledge point mastery evaluation unit and a learning load evaluation unit. The knowledge point mastery evaluation unit analyzes the collected relevant data to evaluate the mastery degree of each knowledge point by the user. The learning load evaluation unit analyzes the collected relevant data to evaluate the current learning pressure of the user; The resource recommendation module adjusts the difficulty of the recommended exercises according to the data obtained from the analysis of the knowledge point mastery evaluation unit. If the user has a good grasp of a certain knowledge point, exercises with a higher difficulty level can be recommended. It also adjusts the knowledge points and exercise order for subsequent learning according to the data obtained from the knowledge point mastery evaluation unit and the learning load evaluation unit, reasonably arranging the knowledge points and exercises to be learned later, making the learning process more scientific and efficient.

[0029] Through the above technical solution, this embodiment provides an intelligent learning resource recommendation system based on data analysis. The system widely collects various aspects of learning data through the behavior collection module, accurately evaluates the mastery of knowledge points and learning pressure through the real-time analysis module, and the resource recommendation module flexibly adjusts the exercise difficulty and subsequent learning order according to the analysis results. The learning resource recommendation is personalized and adapted to the user, accurately recommending appropriate knowledge points and exercises to the user, achieving the effect of significantly improving learning efficiency and pertinence.

[0030] In one embodiment, the process of evaluating the user's mastery of each knowledge point includes:

[0031]

[0032]

[0033] The mastery evaluation coefficient of each knowledge point is obtained through analysis and calculation using formulas (1)-(3) ; Among them, , are the first weight coefficients, reflecting the relative importance of each factor to the mastery evaluation coefficient, which can be preset through experience. is the number of playback times during the learning process of the learning video corresponding to the knowledge point. The number of playback times reflects the difficulty of the user's understanding of the video content. A large number of playback times may indicate difficulty in understanding, and it is obtained by the collection module collecting the user's video learning behavior. is the playback decay coefficient, which can be obtained by fitting experimental data. It is a coefficient that gradually reduces the influence of each playback on the knowledge point mastery evaluation as the number of playback times increases. is the duration influence factor of the learning video corresponding to the knowledge point, which can be obtained by statistically analyzing the mastery of knowledge points corresponding to learning videos of different durations. is the answering correct rate of the exercises corresponding to the knowledge point, which is the proportion of correct answers when the user does the exercises corresponding to the knowledge point. It can be calculated by the behavior collection module recording the user's exercise answering results. The answering correct rate directly reflects the user's application ability of the knowledge point. It is the number of modifications during the answering process of the exercises corresponding to the knowledge points, which can be obtained by recording the user's answering interaction behavior. A large number of modifications may indicate that the user has insufficient mastery of the knowledge points. It is the modification number influence factor, which represents the influence degree of the modification number on the evaluation of knowledge point mastery and can be obtained through experimental data analysis. It is the actual learning duration of the learning video corresponding to the knowledge point, which is the time actually spent by the user on the video corresponding to the knowledge point. The user's video learning time is recorded by the behavior acquisition module, which reflects the energy the user invests in learning this knowledge point. It is the reference learning duration of the learning video corresponding to the knowledge point, which is the reasonable duration set for learning the video corresponding to the knowledge point according to general circumstances or teaching experience. The comparison between the actual learning duration and the reference learning duration can show the user's learning efficiency for this knowledge point. It is the number of pauses during the learning process of the learning video corresponding to the knowledge point, which is the number of times the user pauses when learning the video corresponding to the knowledge point and is obtained by recording the user's video learning behavior. A large number of pauses may indicate that the user encounters confusion during the learning process. It is the pause influence factor, which can be obtained through experimental data and is used to adjust the influence of the number of pauses on the final evaluation coefficient.

[0034] Through the above technical solutions, this embodiment discloses a method for evaluating the user's mastery of each knowledge point. By comprehensively and carefully considering factors such as the playback, duration, pause situation of the learning video, and the answering situation of the exercises, the mastery evaluation coefficient is analyzed and calculated using formulas (1)-(3), obtaining the benefit of more accurately evaluating the user's knowledge point mastery and achieving the effect of providing a more reliable basis for subsequent recommendations.

[0035] In one embodiment, the process of evaluating the user's current learning pressure includes:

[0036] The user's current learning pressure load index is analyzed and calculated through formula (4). ; Among them, It is the actual time used by the user to complete the previous exercise, and the user's answering time is recorded by the behavior acquisition module. It is the reference answering time corresponding to the previous exercise, which is the reasonable answering time set according to the exercise difficulty and the answering situation of general students. The comparison between the actual answering time and the reference time can reflect the user's feeling of the difficulty of answering the questions. It is the number of times the user modifies the answer during the answering process of a certain number of recent exercises, which is statistically obtained by recording the user's answering interaction behavior. $X$ is the number of exercises skipped by the user during the answering process of a certain number of recent exercises, obtained by recording the user's answering behavior. The number of answer modifications and the number of skipped exercises reflect the difficulty level of the user during the answering process. $Y$ is the influence factor of skipped exercises, determined by experimental data, and is used to adjust the role of the number of skipped exercises in the final index. $Z$ is the preset reference number of times, obtained by presetting according to experience. $M$ is the number of times the user switches knowledge points not according to the system-recommended path during the learning process, indicating the number of times the user switches knowledge points not in the order recommended by the system during the learning process, which can be obtained by recording the user's learning path. Switching knowledge points not according to the recommended path may indicate that the user is confused about the current learning content or has their own learning needs. $N$ is the total number of times the user switches knowledge points during the learning process, which is the cumulative number of times the user switches knowledge points during the entire learning process, and can be obtained by counting the user's learning path. 、 、 $W$ is the second weight coefficient, which is the proportion of each factor when calculating the learning pressure load index, and can be obtained by analyzing a large amount of experimental data.

[0037] Through the above technical solution, this embodiment provides a method for evaluating the current learning pressure of a user. The method comprehensively considers factors such as answering time, number of answer modifications, number of skipped exercises, and knowledge point switching, and uses formula (4) to analyze and calculate the learning pressure load index, accurately evaluating the user's learning pressure.

[0038] In one embodiment, the process of adjusting the difficulty of the recommended exercises includes:

[0039]

[0040] The recommended difficulty is obtained by analyzing and calculating through formulas (5)-(6). ; Among them, $A$ is the initial reference difficulty corresponding to the knowledge point, which is the basic difficulty set according to the characteristics of the knowledge point itself and the requirements of the teaching syllabus, and can be obtained by education experts according to factors such as the complexity and importance of the knowledge point. $B$ is the difficulty adjustment coefficient. 、 、 $C$, $D$, $E$ are the difficulty adjustment slope coefficients, which refer to the rate at which the recommended difficulty changes with the knowledge point mastery evaluation coefficient within different knowledge point mastery evaluation coefficient intervals, and can be obtained by fitting experimental data, and are used to control the change range of the recommended difficulty with the knowledge point mastery situation. is a preset evaluation coefficient for mastering the first knowledge point, is a preset evaluation coefficient for mastering the second knowledge point, , is a threshold standard for the degree of mastery set according to experience. By comparing the user's actual mastery with the preset threshold standard for the degree of mastery, the difficulty of the recommended exercises is dynamically adjusted. When the evaluation coefficient for mastering the knowledge point is higher than , the recommended difficulty will increase accordingly. When it is lower than , the recommended difficulty will decrease accordingly. When the evaluation coefficient for mastering the knowledge point is higher than , the recommended difficulty will increase significantly.

[0041] Through the above technical solution, this embodiment provides a method for adjusting the difficulty of recommended exercises. By cleverly combining the initial benchmark difficulty of the knowledge point and the evaluation coefficient of the user's mastery of the knowledge point, and using formulas (5)-(6) to analyze and calculate the recommended difficulty, the purpose of recommending exercises with appropriate difficulty according to the user's actual level can be achieved. The appropriate difficulty system can effectively improve the learning effect.

[0042] In one embodiment, the process of adjusting the difficulty of the recommended exercises further includes: When the user continuously completes a certain number of exercises with the same difficulty, the difficulty adjustment slope coefficient is corrected according to the user's answering performance;

[0043] The corrected difficulty adjustment slope coefficient is obtained through analysis and calculation using formula (7) ; Among them, is the correct rate of a certain number of completed exercises, which is the proportion of correct answers when the user continuously completes a certain number of exercises with the same difficulty and can be calculated by recording the user's answering results. is a preset basic correct rate, which is a reasonable correct rate that can be set according to the teaching goal and the learning situation of general students. After the user continuously completes a certain number of exercises with the same difficulty, the difficulty adjustment slope coefficient is further corrected according to the answering correct rate. If the correct rate is higher than the preset basic correct rate, it means that the user has a good mastery of the exercises at this difficulty level, and the amplitude of difficulty adjustment can be appropriately increased; otherwise, the amplitude of difficulty adjustment is reduced, making the subsequent exercise difficulty adjustment more flexible and accurate. is the correction amplitude coefficient, which is a coefficient that controls the correction amplitude of the difficulty adjustment slope coefficient and can be obtained through experimental data analysis. is the original difficulty adjustment slope coefficient, , is the modified difficulty adjustment slope coefficient.

[0044] Through the above technical solution, this embodiment provides a method for further adjusting the difficulty of recommended exercises. The method can flexibly correct the difficulty adjustment slope coefficient according to the user's answering performance of continuously completing exercises of the same difficulty, can adjust the exercise difficulty more flexibly and accurately, better adapt to the change of the user's learning ability, and can further effectively improve the learning effect.

[0045] In one embodiment, the process of adjusting the subsequent learning knowledge points and exercise order includes: Based on the knowledge graph topology structure, an initial learning path is generated according to the knowledge point dependency relationship. The knowledge graph shows the logical relationship and dependency relationship between knowledge points. Generating an initial learning path according to these relationships can ensure that users learn in a scientific order and avoid knowledge gaps; The user's current learning pressure load index is compared with the preset learning pressure load index critical value ; If , then

[0046]

[0047]

[0048]

[0049] The adjusted learning path is obtained through analysis and calculation by formulas (8)-(11) ; Among them, is the knowledge point coherence score of the learning path, which is used to measure the rationality of the learning path and ensure the systematic learning of knowledge, is the learning novelty score, which is used to increase the novelty of the learning path and stimulate the user's learning interest, is the dynamic adjustment factor, which is used to make the adjustment of the learning path more flexible and adapt to the learning status of different users, is the average value of the mastery evaluation coefficients of each knowledge point in the learning path, which is obtained by averaging the mastery evaluation coefficients of each knowledge point in the learning path, is the preset learning pressure load index critical value, which is obtained by statistical analysis of the learning effect experiments of students under different learning pressures and is used to judge whether the user's learning pressure is too high, so as to determine what method to use to adjust the learning path, , is the third weight coefficient. When calculating the adjusted learning path, the proportions of the knowledge point coherence score and the learning novelty score can be obtained by setting through a large amount of experimental data, and are used to balance the roles of coherence and novelty in learning path adjustment. is the mastery evaluation coefficient corresponding to the i-th knowledge point in the learning path. is the initial benchmark difficulty corresponding to the i-th knowledge point. is the initial benchmark difficulty corresponding to the (i - 1)-th knowledge point. is the proportion of knowledge points not learned by the user in the learning path, which is calculated by comparing the knowledge points in the learning path with the knowledge points already learned by the user, and is used to control the amount of new knowledge learned when adjusting the learning path. is the degree of repetition of the learning path and the user's historical learning path, which is calculated by comparing the learning path with the user's historical learning path, and is used to avoid excessive repetition of the learning path and improve learning efficiency. is the basic correction factor, which can be obtained by setting through experimental data and is used to optimize the adjustment result of the learning path. is the learning path that maximizes the function output value; when the user's learning pressure is within a suitable range, the adjusted learning path is calculated by comprehensively considering factors such as knowledge point coherence, learning novelty, knowledge point mastery evaluation coefficient, and difficulty through formulas (8)-(11), so that the learning path can not only ensure the coherence of knowledge, but also have a certain degree of novelty, while considering the user's knowledge mastery situation and difficulty adaptation.

[0050] If , then only the knowledge points of are retained in the learning path, and the initial learning path length is reduced to half of the original length. is the preset basic mastery evaluation coefficient of knowledge points, which is preset according to experience. When the user's learning pressure is too high, by reducing the number of knowledge points in the learning path and only retaining the knowledge points with better mastery, the user's learning burden is reduced, and the user's learning fatigue caused by excessive pressure is avoided.

[0051] Through the above technical solution, this embodiment discloses a method for adjusting the order of subsequent learning knowledge points and exercises. The method generates an initial learning path based on a knowledge graph, performs different ways of adjustment according to the comparison result between the learning pressure load index and the critical value, and reasonably arranges the learning order according to the user's learning pressure and knowledge mastery situation, achieving the effects of improving learning efficiency and reducing learning burden.

[0052] The process of evaluating the user's mastery of each knowledge point further includes: Analyzing the mastery evaluation coefficient of knowledge points not learned by the user; Extracting the target knowledge point from the knowledge graph All prerequisite dependent knowledge points

[0053]

[0054] Analyze and calculate the mastery evaluation coefficient of unlearned knowledge points through formula (12) ; Among them, is the mastery evaluation coefficient of the prerequisite dependent knowledge point, which can be calculated through formulas (1)-(3) in the front to obtain the user's mastery evaluation coefficient of the prerequisite dependent knowledge point of the target knowledge point, and is used to infer the mastery possibility of the unlearned knowledge point based on the mastery situation of the prerequisite knowledge point. is the influence weight of the prerequisite dependent knowledge point on the mastery evaluation coefficient of the unlearned knowledge point

[0055] For the knowledge points that the user has not learned, evaluate their mastery evaluation coefficient by considering the mastery situation of their prerequisite dependent knowledge points. Because knowledge is coherent and dependent, if the prerequisite knowledge points are mastered well, it may be more beneficial to the understanding and mastery of the subsequent unlearned knowledge points.

[0055] Through the above technical solution, this embodiment discloses a method for evaluating the mastery degree of unlearned knowledge points by the user. The method extracts prerequisite dependent knowledge points from the knowledge graph and analyzes and calculates the mastery evaluation coefficient of unlearned knowledge points through formula (12), so as to more comprehensively evaluate the user's knowledge mastery situation and provide a more accurate basis for recommending unlearned knowledge points.

[0056] The above has described an embodiment of the present invention in detail, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application shall still fall within the scope covered by the patent of the present invention.

Claims

1. An intelligent learning resource recommendation system based on data analysis, characterized in that, The system includes: A behavior collection module, which is used to collect user video learning data, exercise interaction data, and learning rhythm data in real time; A real-time analysis module, including a knowledge point mastery evaluation unit and a learning load evaluation unit. The knowledge point mastery evaluation unit is used to evaluate the user's mastery of each knowledge point, and the learning load evaluation unit is used to evaluate the user's current learning pressure; A resource recommendation module, which adjusts the difficulty of the recommended exercises according to the data obtained by the analysis of the knowledge point mastery evaluation unit; and adjusts the knowledge points and exercise order for subsequent learning according to the data obtained by the analysis of the knowledge point mastery evaluation unit and the learning load evaluation unit.

2. The intelligent learning resource recommendation system based on data analysis according to claim 1, characterized in that, The process of evaluating the user's mastery of each knowledge point includes: ； ； ； The mastery evaluation coefficient of each knowledge point is obtained through analysis and calculation using formulas (1)-(3). ; Among them, , is the first weight coefficient, is the number of replays during the learning process of the learning video corresponding to the knowledge point, is the replay decay coefficient, is the duration influence factor of the learning video corresponding to the knowledge point, is the answering correct rate of the exercises corresponding to the knowledge point, is the number of modifications during the answering process of the exercises corresponding to the knowledge point, is the modification number influence factor, is the actual learning duration of the learning video corresponding to the knowledge point, is the reference learning duration of the learning video corresponding to the knowledge point, is the number of pauses during the learning process of the learning video corresponding to the knowledge point, is the pause influence factor.

3. The intelligent learning resource recommendation system based on data analysis according to claim 2, characterized in that, The process of evaluating the user's current learning pressure includes: ； The current learning stress load index of the user is obtained through analysis and calculation by formula (4). ; Among them, is the actual time taken by the user to complete the previous exercise, is the reference answering time corresponding to the previous exercise, is the number of times the user modifies the answer during the answering process of a certain number of recent exercises, is the number of exercises actively skipped by the user during the answering process of a certain number of recent exercises, is the influence factor of the skipped exercise, is the preset reference number of times, is the number of times the user switches knowledge points not according to the system-recommended path during the learning process, is the total number of times the user switches knowledge points during the learning process, 、 、 are the second weight coefficients.

4. The intelligent learning resource recommendation system based on data analysis according to claim 3, characterized in that, The process of adjusting the difficulty of the recommended exercises includes: ； ； The recommended difficulty is obtained through analytical calculation using formulas (5)-(6). ; Among them, is the initial baseline difficulty corresponding to the knowledge point, is the difficulty adjustment coefficient, , , are the difficulty adjustment slope coefficients, is the preset first knowledge point mastery evaluation coefficient, is the preset second knowledge point mastery evaluation coefficient.

5. The intelligent learning resource recommendation system based on data analysis according to claim 4, characterized in that, The process of adjusting the difficulty of the recommended exercises also includes: When the user continuously completes a certain number of exercises of the same difficulty, the difficulty adjustment slope coefficient is corrected according to the user's answering performance; ； The corrected difficulty adjustment slope coefficient is obtained through analytical calculation using formula (7). ; wherein, is the correct rate of a certain number of completed exercises, is the preset basic correct rate, is the correction amplitude coefficient.

6. The intelligent learning resource recommendation system based on data analysis according to claim 5, characterized in that, The process of adjusting the knowledge points and exercise order for subsequent learning includes: Based on the topological structure of the knowledge graph, an initial learning path is generated according to the knowledge point dependency relationship; Compare the user's current learning stress load index with the preset critical value of the learning stress load index for comparison; If , then ； ； ； ； The adjusted learning path is obtained through analytical calculation using formulas (8)-(11). ; Among them, is the knowledge point coherence score of the learning path, is the learning novelty score, is the dynamic adjustment factor, is the average value of the mastery evaluation coefficients of each knowledge point in the learning path, is the preset critical value of the learning pressure load index, 、 is the third weight coefficient, is the mastery evaluation coefficient corresponding to the i-th knowledge point in the learning path, is the initial benchmark difficulty corresponding to the i-th knowledge point, is the initial benchmark difficulty corresponding to the (i - 1)-th knowledge point, is the proportion of knowledge points not learned by the user in the learning path, is the degree of repetition between the learning path and the user's historical learning path, is the basic correction factor, is the learning path that makes the function output value reach the maximum; If , only the knowledge points in are retained in the learning path, and the length of the initial learning path is reduced to half of the original length. is the preset evaluation coefficient for basic mastery of knowledge points.

7. An intelligent learning resource recommendation system based on data analysis according to claim 6, characterized in that, The process of evaluating the user's mastery of each knowledge point also includes: Analyzing the mastery evaluation coefficient of the knowledge points that the user has not learned; Extract the target knowledge points from the knowledge graph of all the prerequisite dependent knowledge points ; ； The mastery evaluation coefficient of unlearned knowledge points is obtained through analysis and calculation using formula (12). ; Among them, is the mastery evaluation coefficient of prerequisite dependent knowledge points , is the influence weight of the mastery evaluation coefficient of prerequisite dependent knowledge points on the knowledge points that have not been learned .

Images