CN119128233B - Course recommendation method based on learning track and knowledge graph - Google Patents
Course recommendation method based on learning track and knowledge graph Download PDFInfo
- Publication number
- CN119128233B CN119128233B CN202411622413.3A CN202411622413A CN119128233B CN 119128233 B CN119128233 B CN 119128233B CN 202411622413 A CN202411622413 A CN 202411622413A CN 119128233 B CN119128233 B CN 119128233B
- Authority
- CN
- China
- Prior art keywords
- knowledge
- learning
- knowledge point
- point
- mastery
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9035—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9038—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Educational Technology (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
本发明公开了一种基于学习轨迹与知识图谱的课程推荐方法，具体涉及教育课程推荐技术领域，包括以下步骤：通过收集和整合学生的学习轨迹数据，映射至知识图谱生成学习轨迹－知识联合图谱，进行知识点掌握和学习率分析，得到每个知识点的知识掌握值与学习率；基于知识掌握值，系统将知识点划分为强项和弱项；对于弱项知识点，结合知识掌握值与学习率计算弱项推荐度，并根据推荐度优先推荐课程；对于强项知识点，基于学习率计算强项推荐度，推荐进阶课程，提升学生的学习效果和个性化学习体验；本发明可以精准识别学生的强项和弱项知识点，从而提供针对性极强的学习推荐，使得学生能够在个性化的学习路径上更高效地学习，提升学习成果。
The present invention discloses a course recommendation method based on learning trajectory and knowledge graph, which specifically relates to the technical field of educational course recommendation, and comprises the following steps: by collecting and integrating students' learning trajectory data, mapping to the knowledge graph to generate a learning trajectory-knowledge joint graph, performing knowledge point mastery and learning rate analysis, and obtaining the knowledge mastery value and learning rate of each knowledge point; based on the knowledge mastery value, the system divides the knowledge points into strengths and weaknesses; for weak knowledge points, the weak recommendation degree is calculated in combination with the knowledge mastery value and the learning rate, and courses are recommended preferentially according to the recommendation degree; for strong knowledge points, the strong recommendation degree is calculated based on the learning rate, and advanced courses are recommended to improve students' learning effect and personalized learning experience; the present invention can accurately identify students' strong and weak knowledge points, thereby providing highly targeted learning recommendations, so that students can learn more efficiently on a personalized learning path and improve learning outcomes.
Description
Technical Field
The invention relates to the technical field of education course recommendation, in particular to a course recommendation method based on learning tracks and knowledge maps.
Background
Course recommendation based on learning tracks and knowledge patterns is an intelligent education recommendation mode, and the courses are recommended in a personalized mode by combining learning historical data of students and the knowledge patterns. First, the learning track refers to a behavior record left by a student in the learning process, including learning progress, learning content, examination results, etc., and by analyzing these data, learning habits and knowledge mastering conditions of the student can be known. The knowledge graph is a network representing the association relationship between knowledge points, and the knowledge graph is used for helping to reveal the knowledge points on which students have defects or need to learn further through logic connection of the knowledge points. By mapping the learning track of the student to the knowledge graph, learning weaknesses of the student and the knowledge field needing to be improved can be found.
With rapid development of education technology, personalized learning is becoming an important trend of online education. The traditional teaching method often cannot accurately identify knowledge mastering conditions and learning requirements of each student, so that course recommendation is rough, and personalized learning requirements of different students are difficult to meet. Therefore, a course recommendation method based on learning trajectories and knowledge maps is provided herein.
Disclosure of Invention
In order to achieve the above purpose, the present invention provides the following technical solutions:
A course recommendation method based on learning tracks and knowledge maps comprises the following steps:
collecting learning track data of students and integrating the learning track data to obtain a learning track data integration set of the students;
constructing a knowledge graph, decomposing course content into knowledge points, and mapping a learning track data integration set of students onto the knowledge graph to obtain a learning track-knowledge combined graph;
Carrying out knowledge point mastering analysis based on the learning track-knowledge combined map to obtain knowledge mastering values of each knowledge point, and simultaneously carrying out knowledge point learning analysis to obtain knowledge learning rate of each knowledge point;
Judging knowledge point grasping degree based on the knowledge grasping value, and dividing all knowledge points into a strong knowledge point set and a weak knowledge point set;
for the weak item knowledge point set, obtaining weak item recommendation degree according to knowledge mastering values and knowledge learning rates, and performing course recommendation according to the weak item recommendation degree;
and for the strong knowledge point set, the knowledge learning rate of the target knowledge point is strong recommendation degree, and course recommendation is carried out according to the strong recommendation degree.
In a preferred embodiment, knowledge point grasping analysis is performed based on the learning track-knowledge combination map, and obtaining a knowledge grasping value of each knowledge point means:
respectively acquiring and marking the total learning time spent by the students on the target knowledge point i The review frequency of students on the target knowledge point i is marked asThe accuracy of the j-th exercise of the student on the target knowledge point i is marked asTest performance of the kth test of the student on the target knowledge point i is marked asThen substituting the knowledge mastery value into a knowledge mastery value calculation formula of the target knowledge point i:
;、、 Are all preset influence coefficients 、、The sum of which is one,Is a positive value, and is a positive value,The value of the slope parameter of the logic Stir function is in the range of 0.01 to 10,Is the review frequencyA corresponding preset threshold value, n is the total number of exercises, m is the total number of exams,Preset adjustment coefficients for the jth exercise,For the adjustment coefficient of the kth test,The knowledge of the target knowledge point i is grasped.
In a preferred embodiment, a student learning model is built based on knowledge mastery values:
acquiring knowledge mastery value of student at initial time point on target knowledge point i and marking as The maximum mastery value increment which can be finally reached by the student is marked asThe following models were built:
to minimize the predicted value For the purpose of determining the attenuation coefficient from the actual value differenceIs a function of the number of (c),Knowledge grasping value representing corresponding prediction at time t, attenuation coefficientThe determined numerical value is substituted into a student learning model for application.
In a preferred embodiment, knowledge point learning analysis is performed, and obtaining a knowledge learning rate of each knowledge point means:
the method comprises the steps of obtaining a student learning model applied by a student on a target knowledge point i, and then obtaining the knowledge learning rate of the student on the target knowledge point i:
; Representing the learning rate of the student's knowledge at the target knowledge point i.
In a preferred embodiment, the knowledge point grasping degree judgment based on the knowledge grasping value means that:
And respectively acquiring knowledge mastering values of the students on each knowledge point, comparing the knowledge mastering values with a preset knowledge point mastering degree threshold, marking the knowledge point as a learning strong knowledge point if the knowledge mastering value on the knowledge point is greater than or equal to the preset knowledge point mastering degree threshold, and marking the knowledge point as a learning weak knowledge point if the knowledge mastering value on the knowledge point is less than the preset knowledge point mastering degree threshold.
In a preferred embodiment, the division of all knowledge points into strong knowledge point sets and weak knowledge point sets refers to:
summarizing all the knowledge points marked as corresponding to learning strong knowledge points to obtain a strong knowledge point set, and summarizing all the knowledge points marked as corresponding to learning weak knowledge points to obtain a weak knowledge point set.
In a preferred embodiment, obtaining the weak item recommendation degree according to the knowledge mastery value and the knowledge learning rate means:
mapping the knowledge mastering value and the knowledge learning rate into the numerical ranges of [0,1] respectively to obtain mapped knowledge mastering values Knowledge learning rateThen substituting the weak term recommendation degree into a calculation formula of the weak term recommendation degree:
; for the degree of recommendation of the weak item, 、Are both preset recommended coefficients and the sum of the two is one.
In a preferred embodiment, course recommendation according to weak item recommendation level refers to:
Recommendation degree of weak item And performing descending order arrangement to obtain weak item recommendation main page sheets, clicking knowledge points on each weak item recommendation main page sheet to enter weak item recommendation auxiliary page sheets corresponding to the knowledge points, wherein each weak item recommendation auxiliary page sheet consists of a plurality of courses containing the knowledge points, and the courses containing the knowledge points are arranged in descending order according to the content similarity of the courses and the knowledge points.
The invention has the technical effects and advantages that:
By analyzing the learning track and the knowledge graph of the student, the invention can customize a personalized learning path for the student. According to the knowledge mastering value and the learning rate of the students, strong knowledge points and weak knowledge points of the students can be accurately identified, so that highly-targeted learning recommendation is provided, the students can learn more efficiently on personalized learning paths, and learning results are improved.
The invention can dynamically adjust recommended learning content and course difficulty, and provides the most suitable learning material through comprehensive analysis of learning track, knowledge mastering value and learning rate. The learning efficiency is improved, students are helped to master knowledge more quickly, strong items are further consolidated and improved, and weak links are enhanced. By continuously updating and tracking the learning track of the student, the invention can feed back the learning state of the student in real time. Both the bottleneck in learning and the progress can be recognized and adjusted in time, thereby helping students dynamically optimize personal learning strategies.
According to the invention, through comprehensive analysis of knowledge mastering values and knowledge learning rates, weak knowledge points and strong knowledge points of students can be accurately evaluated. For weak knowledge points, the system can recommend related courses preferentially to help students to make up knowledge holes, and for strong knowledge points, the system recommends higher-order advanced courses to further strengthen knowledge reserve and application capacity of the students in the fields. According to the invention, learning resources can be reasonably allocated through intelligent knowledge point analysis and course recommendation, and unnecessary resource waste is avoided. The time and energy of students can be concentrated in the field which needs to be lifted most, and the learning effect is improved to the greatest extent.
Drawings
For the convenience of those skilled in the art, the present invention will be further described with reference to the accompanying drawings;
Fig. 1 is a schematic diagram of a course recommendation method based on learning trajectories and knowledge maps in the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The following examples are obtained with reference to fig. 1:
embodiment 1 a course recommendation method based on learning trajectories and knowledge maps, comprising the following steps:
And the learning track data reflects the learning behavior and learning process of the students and is the basis for learning the learning habit and knowledge mastering condition of the students. Such data may include student's learning time, number of exercises, accuracy, examination results, etc. The learning track data can be derived from different learning systems and platforms, and the integrated process is not only limited to simple data combination, but also data cleaning and normalization processing. The ensemble may contain learning activities including, but not limited to, duration of learning (length of time per learning), performance of exercise (frequency and accuracy of exercise), review (number and interval of review), examination results (examination results associated with a particular knowledge point).
And constructing a knowledge graph, decomposing course content into knowledge points, and mapping a learning track data integration set of the students onto the knowledge graph to obtain a learning track-knowledge combination graph, wherein the knowledge graph is a structural representation of the knowledge points and the relationship between the knowledge points, so that the students can understand the dependency relationship between the knowledge points. The learning track data of the students are mapped to the knowledge graph, so that the students can more clearly see which knowledge points are excellent in performance and which knowledge points still need to be enhanced. The process of constructing the knowledge graph not only comprises the decomposition of the knowledge points, but also relates to the association (such as pre-knowledge and follow-up knowledge) among the knowledge points. Through the mapping, a map combining the learning behaviors and the knowledge structure of the students can be generated, which is called a learning track-knowledge combined map.
And carrying out knowledge point mastering analysis based on the learning track-knowledge combined map to obtain a knowledge mastering value of each knowledge point, and carrying out knowledge point learning analysis to obtain a knowledge learning rate of each knowledge point, wherein the knowledge mastering value is an index for measuring the mastering condition of a student on a specific knowledge point and is calculated based on the learning track of the student. The knowledge learning rate is the speed or efficiency of the student when learning a certain knowledge point, and reflects the learning ability of the student on the knowledge point. The calculation of the knowledge mastering value may integrate a plurality of factors such as learning time, exercise accuracy, review frequency and examination results. The calculation of the knowledge learning rate can then be measured by the variation of the knowledge mastery value over different time periods. From these analyses, it can be identified which knowledge points the student has better and which still requires more time and effort.
Judging knowledge point grasping degree based on the knowledge grasping value, and dividing all knowledge points into a strong knowledge point set and a weak knowledge point set; by classifying the knowledge mastery values of students, knowledge points of students can be classified into strong items and weak items. Therefore, students can learn more pertinently, weak knowledge points are strengthened, and strong knowledge points are consolidated. The classification of knowledge points may be based on the interval of knowledge mastery values, and different level criteria may be set (for example, mastery value >0.7 is a strong term and <0.4 is a weak term). The process not only can help students identify own learning weak links, but also can provide feedback on the learning state of the students for teachers.
And the weak knowledge points need to be processed preferentially so as to help students to make up for the loopholes in learning. The weak item recommendation degree combines the knowledge mastering value and the learning rate, and preferably recommends courses corresponding to knowledge points with lower student mastering degree and slower learning rate. The weak term recommendation degree can be calculated through a formula, and knowledge points with lower knowledge mastery values and slower learning rates are given higher recommendation weights. The recommended course content is more specific to the weak knowledge points, and the course content which is highly overlapped with the knowledge points can be recommended preferentially. Meanwhile, the recommended course can help students to better master weak knowledge by adjusting teaching difficulty and providing more training and review opportunities.
For the strong knowledge point set, the knowledge learning rate of the target knowledge point is strong recommendation degree, and course recommendation is carried out according to the strong recommendation degree, wherein the strong knowledge point represents knowledge which students have better to master, but still needs to be consolidated and promoted. The strong courses are recommended according to the knowledge learning rate, so that the learning efficiency of students can be further improved, and the students can obtain higher learning effect on the good knowledge points. The degree of strong item recommendation is determined mainly according to the learning rate of students. Knowledge points with higher learning rate can recommend deeper and more complex course content, and help students further expand relevant knowledge or perform advanced application. Recommended courses may cover knowledge points of expansibility, helping students to get further on the basis of existing strong items.
Through the steps, not only can the learning condition of the students be accurately analyzed, but also courses can be recommended for the students according to personalized weak items and strong items. The mode improves the pertinence and the efficiency of learning, so that students can better overcome weak links in the learning process, and consolidate and promote strong items.
Knowledge point mastering analysis is carried out based on the learning track-knowledge combined map, and knowledge mastering values of each knowledge point are obtained by the following steps:
respectively acquiring and marking the total learning time spent by the students on the target knowledge point i The review frequency of students on the target knowledge point i is marked asThe accuracy of the j-th exercise of the student on the target knowledge point i is marked asTest performance of the kth test of the student on the target knowledge point i is marked asThen substituting the knowledge mastery value into a knowledge mastery value calculation formula of the target knowledge point i:
;、、 Are all preset influence coefficients 、、The sum of which is one,Is a positive value, and is a positive value,The value of the slope parameter of the logic Stir function is in the range of 0.01 to 10,Is the review frequencyA corresponding preset threshold value, n is the total number of exercises, m is the total number of exams,Preset adjustment coefficients for the jth exercise,For the adjustment coefficient of the kth test,The knowledge of the target knowledge point i is grasped. When the target knowledge pointsThe greater the value, the higher the student's level of mastery of the knowledge point. That is, the students have more time, higher exercise accuracy and review frequency in the process of learning the knowledge points, and the related examination results have good performance and larger performanceThe value represents that the learning effect of the student on the knowledge point is better and the student can master more firmly.
Indicating the accuracy of the student at knowledge point i during the j-th exercise. Since students may perform multiple exercises, the accuracy of each exercise may be different, and thus a summary or weighting process may be required for the multiple exercises. j represents the number of exercises and is combined with the preset adjustment coefficient of the jth exerciseA weighting process may be performed to obtain the student's practice performance at the knowledge point.Representing the student's performance at knowledge point i at the kth examination. This suggests that multiple test achievements for a particular knowledge point need to be combined together to reflect the student's overall performance in different tests. k represents the number of the test and is combined with the adjustment coefficient of the kth testThe examination performance of students on the knowledge points can be comprehensively measured.
Total learning timeThe impact of (a) does not grow linearly and its marginal benefit decreases with increasing time. Therefore, the learning time is processed by using a logarithmic function, and overestimation of the learning effect after long-time investment is avoided. The logarithmic function may well mimic this decrementing effect.
The influence of the exercise accuracy is corrected by the review frequency, and when the review frequency is higher, the grasping effect of students in the exercise is more obviously improved, so that the influence of the accuracy is enhanced along with the increase of the review frequency. However, in order to avoid the effect of excessively depending on the review frequency, the square term in the denominator is used to control the influence of the review frequency.
The review frequency affects the performance of the exercise and examination results, so that the effect of the review frequency is reflected by the exercise accuracy and examination results. The higher the review frequency, the better the training effect is theoretically, and the better the examination performance is, but the effect is gradually stable along with the increase of the review frequency.
The impact of the test performance is dynamically adjusted by review frequency. Using a logistic function when the review frequency is above a certain thresholdIn this case, the weight of the examination score is significantly increased, and otherwise the effect is weaker. The form reflects the 'starting' effect of the review on the examination results, namely, the examination results are greatly improved after students need to reach a certain review frequency.
、、Is used for balancing the contribution of learning time, exercise accuracy and examination results to knowledge mastery values. Can be adjusted according to course characteristics.The influence degree of the review frequency on the exercise accuracy is controlled to be positive. The larger the value, the more pronounced the effect of review frequency.And controlling the rate of improvement of the review frequency on the examination score as the slope parameter of the logic Stir function. The larger the value, the more obvious the review frequency affects the performance of the examination.The value of (2) is typically a positive number because it controls the rate and must be positive to ensure logical rationality of the model.The value of (2) affects the sensitivity of the model, the larger the value, the more rapid the response, and the smaller the value, the more gradual the response. The value range of 0.01 to 10 is a common value range, and specific values can be set according to the scale of data and the change requirement of learning rate.
It should be noted that, in the calculation formula of the knowledge mastery value, the logarithm base may be a natural logarithm (base is e) or a common logarithm (base is 10), which depends on the actual application scenario and the desired effect. Natural logarithms (base e) are commonly used for scientific calculations or mathematical modeling, as they are closely related to the exponential growth process, especially for marginal effect decrementing representing learning behavior. Natural logarithms are very often used when dealing with models of exponential growth, decay, etc., and are suitable for situations where marginal gains representing learning time decrease. A common logarithm (base 10) may be used if a more intuitive result is desired or more practical in some applications. Common logarithms are commonly used for information content analysis, and are suitable for use when the data size is large or when the results are desired to be presented in a smaller range of values. Because of the phenomenon of decreasing marginal effect of learning behavior, the invention hopes that the model reflects more naturally the nonlinear growth or decay in the learning process, and the natural logarithm (base e) can be selected. If the user wants the result to be more visual and concise in a certain range in actual use, the user can select the common logarithm (base 10).
Establishing a student learning model based on the knowledge mastery value, namely acquiring the knowledge mastery value of the student at the initial time point on the target knowledge point i and marking the knowledge mastery value asThe maximum mastery value increment which can be finally reached by the student is marked asThe following models were built: to minimize the predicted value For the purpose of determining the attenuation coefficient from the actual value differenceIs a function of the number of (c),Knowledge grasping value representing corresponding prediction at time t, attenuation coefficientThe determined numerical value is substituted into a student learning model for application. By establishing a dynamic change model of knowledge mastering values, knowledge mastering conditions of students on a certain knowledge point can be estimated according to time. The model not only considers the knowledge mastering level of students at the beginning of a certain knowledge point, but also reflects the mastering improvement of the students along with the time learning through an exponential decay formula, and finally tends to a certain maximum value. The exponential decay factor controls the learning speed, i.e. the learning time is later, the learning speed is gradually slowed down.
The knowledge point learning analysis is carried out, and the knowledge learning rate of each knowledge point is obtained by acquiring a student learning model applied by a student on a target knowledge point i, and then the knowledge learning rate of the student on the target knowledge point i:; Representing the learning rate of the student's knowledge at the target knowledge point i. The learning rate of the students on the target knowledge points is obtained through the derivation of the formulas, and the learning efficiency of the students at different time points is described. With this rate, the teacher can identify bottlenecks and progress in the student's learning. In the expression of the learning rate, the rate is in direct proportion to the maximum mastering increment of the students and the learning attenuation parameter, which indicates that the learning speed in the initial stage is higher, and the learning effect gradually becomes stable along with the time. The model can provide basis for personalized learning plans, can dynamically adjust recommended learning content according to different learning rates and mastering conditions of students, can track learning progress and mastering states of the students in real time, and helps teachers and students to better learn weak links in the learning process. For educational institutions and teachers, the model is helpful for identifying whether learning efficiency of students at certain knowledge points is low in the teaching process, and then targeted coaching and intervention are performed.
The step of judging the knowledge point mastering degree based on the knowledge mastering values refers to the steps of respectively acquiring the knowledge mastering value of each knowledge point of a student and comparing the knowledge mastering value with a preset knowledge point mastering degree threshold value, marking the knowledge point as a learning strong knowledge point if the knowledge mastering value of the knowledge point is greater than or equal to the preset knowledge point mastering degree threshold value, and marking the knowledge point as a learning weak knowledge point if the knowledge mastering value of the knowledge point is less than the preset knowledge point mastering degree threshold value. Summarizing all the knowledge points marked as corresponding to learning strong knowledge points to obtain a strong knowledge point set, and summarizing all the knowledge points marked as corresponding to learning weak knowledge points to obtain a weak knowledge point set. By classifying strong items and weak items of knowledge points according to the level of knowledge mastering values, a learning system or a teacher can be helped to customize a personalized learning path for students. Strong knowledge points can reduce review, while weak knowledge points require important attention and learning. This way, learning efficiency can be improved, enabling students to more effectively allocate learning time. Knowledge mastery values can be updated continuously as time progresses and as learning processes progress. When the knowledge point mastery degree of the weak item of the student is improved, the learning strategy can be dynamically adjusted, so that the student can be helped to consolidate the knowledge point in time or overcome the new weak item knowledge point, and efficient knowledge point management is realized.
Obtaining weak item recommendation degree according to knowledge mastering value and knowledge learning rate refers to mapping the knowledge mastering value and knowledge learning rate into numerical ranges of [0,1] respectively to obtain mapped knowledge mastering valueKnowledge learning rateThen substituting the weak term recommendation degree into a calculation formula of the weak term recommendation degree:
; for the degree of recommendation of the weak item, 、All are preset recommendation coefficients, the influence of knowledge mastering values and learning rates on weak item recommendation degrees is measured respectively, and the sum of the knowledge mastering values and the learning rates is one.The mapping result of the knowledge mastering value represents the mastering degree of the knowledge point by the student, and the larger the mapping result is, the better the mastering is.The mapping result of the knowledge learning rate represents the learning efficiency of the student on the knowledge point, and the larger the learning efficiency is, the higher the learning efficiency is. The greater the weak item recommendation, the greater the degree of weakness that represents the knowledge point, meaning that the student is not adequately aware of the knowledge point or learning is less efficient, and therefore, needs to recommend the relevant course preferentially.
Course recommendation based on weak item recommendation level refers to recommendation level of weak itemAnd performing descending order arrangement to obtain weak item recommendation main page sheets, clicking knowledge points on each weak item recommendation main page sheet to enter weak item recommendation auxiliary page sheets corresponding to the knowledge points, wherein each weak item recommendation auxiliary page sheet consists of a plurality of courses containing the knowledge points, and the courses containing the knowledge points are arranged in descending order according to the content similarity of the courses and the knowledge points. The purpose of weak item recommendation degree ranking is to rank the priority recommendation orders of knowledge points according to the weak item recommendation degree calculated previously. The higher the weak item recommendation, the weaker the knowledge point, so the corresponding course will be recommended preferentially. The user can click on a knowledge point in the weak term recommendation main page to enter the weak term recommendation auxiliary page list. And displaying a plurality of courses related to the knowledge point in the auxiliary page, and sorting according to the similarity of the courses and the content of the knowledge point. Thus, students can preferentially see courses most relevant to weak knowledge points of students, and study the courses in a targeted manner.
For example, supposing that the weak item knowledge point of a certain student is the chain rule in calculus, the weak item recommendation degree of the knowledge point is higher through analysis of knowledge mastering values and learning rates. Therefore, "chain rule" is used as an important recommendation in the weak item recommendation main page. When the student clicks on the knowledge point, he enters a sub-page, presenting a plurality of chain rule related courses, which may include a base course, a definition of introducing the chain rule and a basic application. Middle class course, explaining the application of chain rule in different mathematic scenes in detail. Advanced course-advanced application of the chain law in multivariate calculus. The courses are ordered according to the similarity of the content and the chain rule, so that the most relevant courses are ensured to be arranged in front, and students are helped to find the most suitable learning resources quickly. Course resources are typically from cloud-based course libraries that store large amounts of structured learning materials that can be retrieved by keywords and knowledge points. The cloud course library can be a self-built resource of the education platform, and can also be integrated by a third-party course provider. These resources may include courses of varying difficulty levels, covering content from basic to advanced. Similarity of course and knowledge point content can be achieved through various prior art technologies, and common algorithms include similarity calculation (such as TF-IDF or Word2 Vec) based on Word vectors, wherein the similarity is judged by converting course content and knowledge point description into Word vectors and calculating cosine similarity between the course content and the knowledge point description. The specific similarity calculation process is not described in detail herein.
For the strong knowledge point set, the knowledge learning rate of the target knowledge point, namely the strong recommendation degree, and performing course recommendation according to the strong recommendation degree are executed according to a preset strategy, specifically, in the strong knowledge point set, the knowledge learning rate is an index for measuring the learning efficiency of students on the knowledge points. The learning rate of strong knowledge points is generally higher, which indicates that students learn on the knowledge points quickly and grasp the knowledge points firmly. Therefore, the learning rate can be directly used as the strong item recommendation degree. And descending order arrangement is carried out on the strong knowledge points according to the calculated strong recommendation degree, and higher-order courses corresponding to the knowledge points with highest learning rate are recommended preferentially. The higher the strong recommendation, the more solid the student is in grasp, and the faster the learning speed, so more challenging courses are recommended. In the cloud course library or the internal course library of the learning system, a pre-marked course difficulty label exists. By matching the knowledge point correlation of the strong knowledge points with the knowledge points of the higher-order courses, suitable higher-order courses can be recommended for students. Assuming that the student has a high knowledge learning rate at the knowledge point "differential equation", the knowledge point can be listed as a strong term. According to the learning condition of students on the differential equation, advanced courses such as the high-difficulty courses of the partial differential equation and the application of the differential equation in physics can be recommended, and the capability of the students in the field is further improved. Higher-order and more challenging courses are recommended to students, and students are guaranteed to be further promoted in the field of the strong items. The method can help students expand knowledge surface on the basis of existing advantages, and larger learning potential is mined.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
It should be understood that, in various embodiments of the present application, the sequence numbers of the foregoing processes do not mean the order of execution, and the order of execution of the processes should be determined by the functions and internal logic thereof, and should not constitute any limitation on the implementation process of the embodiments of the present application.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (5)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411622413.3A CN119128233B (en) | 2024-11-14 | 2024-11-14 | Course recommendation method based on learning track and knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411622413.3A CN119128233B (en) | 2024-11-14 | 2024-11-14 | Course recommendation method based on learning track and knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119128233A CN119128233A (en) | 2024-12-13 |
| CN119128233B true CN119128233B (en) | 2025-01-24 |
Family
ID=93769144
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411622413.3A Active CN119128233B (en) | 2024-11-14 | 2024-11-14 | Course recommendation method based on learning track and knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119128233B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119416875B (en) * | 2025-01-07 | 2025-04-08 | 辽宁蒲公英教育咨询有限公司 | High-order question chain generation method and system based on standardized knowledge graph construction |
| CN119540007B (en) * | 2025-01-21 | 2025-05-02 | 山东顺势教育科技集团有限公司 | Personalized semantic evaluation system combined with knowledge graph |
| CN119905024B (en) * | 2025-04-01 | 2025-07-11 | 浙江海亮科技有限公司 | Learning assistance method and device, storage medium and electronic equipment |
| CN119919013A (en) * | 2025-04-02 | 2025-05-02 | 漳州城市职业学院 | A method and system for evaluating students' quality and accomplishment |
| CN120634812A (en) * | 2025-08-12 | 2025-09-12 | 中亿丰数字科技集团股份有限公司 | Intelligent education platform course recommendation method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113851020A (en) * | 2021-11-04 | 2021-12-28 | 华南师范大学 | Self-adaptive learning platform based on knowledge graph |
| CN114065031A (en) * | 2021-11-12 | 2022-02-18 | 云南师范大学 | A personalized learning path recommendation method based on fuzzy cognitive graph |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6554618B1 (en) * | 2001-04-20 | 2003-04-29 | Cheryl B. Lockwood | Managed integrated teaching providing individualized instruction |
| US10546507B2 (en) * | 2014-09-18 | 2020-01-28 | International Business Machines Corporation | Recommending a set of learning activities based on dynamic learning goal adaptation |
| CN118036774A (en) * | 2024-03-19 | 2024-05-14 | 江苏师范大学 | A personalized learning recommendation system based on federated learning |
-
2024
- 2024-11-14 CN CN202411622413.3A patent/CN119128233B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113851020A (en) * | 2021-11-04 | 2021-12-28 | 华南师范大学 | Self-adaptive learning platform based on knowledge graph |
| CN114065031A (en) * | 2021-11-12 | 2022-02-18 | 云南师范大学 | A personalized learning path recommendation method based on fuzzy cognitive graph |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119128233A (en) | 2024-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN119128233B (en) | Course recommendation method based on learning track and knowledge graph | |
| Brahim | Predicting student performance from online engagement activities using novel statistical features | |
| Biehler et al. | Introducing students to machine learning with decision trees using CODAP and Jupyter Notebooks | |
| CN109858797B (en) | Multi-dimensional informatics analysis method based on knowledge network accurate online education system | |
| CN111651676B (en) | Method, device, equipment and medium for performing occupation recommendation based on capability model | |
| CN117172977A (en) | Training suggestion generation method and system for trainee training | |
| CN119761524A (en) | Recommendation methods and related equipment for artificial intelligence teaching resources in primary and secondary schools | |
| Drăgulescu et al. | Predicting assignment submissions in a multi-class classification problem | |
| CN115249072B (en) | A reinforcement learning path planning method based on generative adversarial user model | |
| Costa et al. | Automatic classification of computational thinking skills in elementary school math questions | |
| Pan et al. | Dynamic Task Allocation Algorithm for Hiring Workers that Learn. | |
| CN114943625A (en) | Remote education system based on block chain technology | |
| CN112766765A (en) | Professional learning ability evaluation method and system based on interval middle intelligence theory | |
| OROZOVA et al. | Generalized net model for dynamic decision making and prognoses | |
| CN114357297B (en) | Student portrait construction and learning resource distribution method, computer equipment and storage medium | |
| Bourdache et al. | Active preference elicitation by bayesian updating on optimality polyhedra | |
| Petrov | Multi-criteria Evaluation of Students’ Performance Based on Hybrid AHP-Entropy Approach with TOPSIS, MOORA and WPM | |
| Etaat | AI-based assessment of students’ writing skill progress: Implementing “Artificial neural network modeling” vs.“Time series prediction” | |
| JP7439914B2 (en) | Prediction device, prediction method, and program | |
| Capuano et al. | Multi-criteria fuzzy ordinal peer assessment for MOOCs | |
| Boschetti et al. | Interactive modelling for natural resource management | |
| Caixue | Design of a Multi-Indicator Evaluation System for Art Students’ Learning Outcomes Driven by Neural Networks: Weight Optimization Using the Entropy Method | |
| Williams | Markov Chain Monte Carlo Methods for Dynamic Student Assessment | |
| CN117726082B (en) | Teaching resource recommendation method and device, electronic equipment and readable storage medium | |
| Shou et al. | All Element Selection Method in Classroom Social Networks and Analysis of Structural Characteristics. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |