An online course learning recommendation method based on a knowledge graph

Abstract

The present application relates to the technical field of online course learning recommendation, in particular to an online course learning recommendation method based on a knowledge graph. The method can improve the use value of the obtained data by preprocessing the obtained user information and online course information, and can make the knowledge graph better label these data by inputting the preprocessed user information and online course information into the knowledge graph. The method can make the online course recommendation more accurate by calculating the adaptation degree of the user and the course package based on the label and recommending the course package to the user according to the adaptation degree. In addition, the method can determine whether the recommendation result meets the standard according to the browsing volume of each recommended online course, and preliminarily determine the reason why the standard is not met, so that the online course can be pushed more accurately according to the user's demand, the efficiency of the user in the course learning process can be improved, and the efficiency of the online course recommendation can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of online course learning recommendation, in particular to an online course learning recommendation method based on a knowledge graph. BACKGROUND

[0002] The online course learning recommendation method based on a knowledge graph is a new recommendation method combining knowledge graph technology and recommendation algorithms. This method constructs a course knowledge graph by analyzing learners' behavior data on online course platforms, course attributes, and the association between courses, and provides personalized course recommendations for learners based on the graph, so this method provides more accurate course recommendations for learners and can tailor personalized course recommendation plans for them according to their personal circumstances and interest preferences.

[0003] Chinese Patent Publication No. CN116501970A discloses an online course recommendation method based on a knowledge graph and convolution, which includes: the model first uses a feature extraction module to extract and convert user historical interaction information and course domain feature information into embedding vectors, and fuses these vectors into a knowledge graph to generate user and course embedding expressions, and finally transmits these information to adjacent nodes through a message propagation algorithm to obtain the final embedding vectors of users and courses for recommendation.

[0004] As can be seen, the above-mentioned scheme has the following problems: it fails to adjust the recommended online courses in a timely manner according to the user's needs, resulting in low efficiency of the user in the course learning process, and thus reducing the efficiency of online course recommendation. SUMMARY

[0005] To this end, the present application provides an online course learning recommendation method based on a knowledge graph to overcome the problem in the prior art that the recommended online courses are not adjusted in a timely manner according to the user's needs, resulting in low efficiency of the user in the course learning process, and thus reducing the efficiency of online course recommendation.

[0006] To achieve the above-mentioned purpose, the present application provides an online course learning recommendation method based on a knowledge graph, comprising:

[0007] obtaining user information and online course information;

[0008] preprocessing the user information and the online course information respectively;

[0009] inputting the preprocessed user information and online course information into a knowledge graph, the knowledge graph packs the online courses into a number of course packages according to their features, outputs a number of preset labels for each course package and a number of demand labels for the user according to the user information, wherein the preset labels are the common features of all courses in the course package;

[0010] calculating a degree of adaptation of the course package and the user demand, wherein the degree of adaptation is a ratio of a number of same tags in preset tags of the course package and tags of the user demand to a total number of the tags of the user demand;

[0011] recommending the course package to the user according to the degree of adaptation, and pushing the course package to the user when the degree of adaptation is higher than a preset degree of adaptation;

[0012] determining whether the recommendation result of the online course meets a standard based on a browsing volume of each recommended course package, and determining a corresponding processing mode based on a reason why the recommendation result does not meet the standard;

[0013] adjusting the online course learning recommendation method based on the processing mode.

[0014] Further, the process of determining whether the recommendation result of the online course meets the standard based on the browsing volume of each course package includes: determining whether the recommendation result meets the standard based on a comparison result of the browsing volume and a preset browsing volume stored in advance; determining that the recommendation result meets the standard if the browsing volume is greater than or equal to a first preset browsing volume; determining whether the recommendation result meets the standard based on a cumulative learning duration of the user for the recommended course package if the browsing volume is less than the first preset browsing volume and greater than a second preset browsing volume, wherein the cumulative learning duration is an average duration of learning the course by the user receiving the course package; and determining that the recommendation result does not meet the standard if the browsing volume is less than or equal to the second preset browsing volume, and determining a reason why the recommendation result does not meet the standard based on a difference between a browsing volume at a previous time and a current browsing volume.

[0015] Further, the process of determining whether the recommendation result meets the standard based on the cumulative learning duration of the user for the recommended course package includes: comparing the cumulative learning duration with a preset learning duration; determining that the recommendation result does not meet the standard if the cumulative learning duration is less than or equal to the preset learning duration, and determining a reason why the recommendation result does not meet the standard based on a difference between a browsing volume at a previous time and a current browsing volume; and determining that the recommendation result meets the standard if the cumulative learning duration is greater than the preset learning duration, and adjusting the preset browsing volume based on an actual usage ratio, wherein the actual usage ratio is a ratio of a number of users whose cumulative learning duration is greater than the preset learning duration to a total number of users receiving the push of the course package.

[0016] Further, the process of adjusting the preset browsing volume based on the actual usage ratio includes: calculating a difference between the actual usage ratio and a preset usage ratio; and reducing the preset browsing volume based on the calculated difference, and the reduction amplitude of the preset browsing volume is proportional to the difference.

[0017] Further, the process of adjusting the preset browsing amount further comprises: calculating a difference between the average interaction ratio of the user and the online course and the preset interaction ratio; and reducing the preset browsing amount based on the obtained difference, and the difference is proportional to the reduction amplitude of the browsing amount.

[0018] Further, the process of determining the reason why the recommendation result does not meet the standard when the browsing amount is less than or equal to the second preset browsing amount comprises: calculating an absolute value of a difference between the browsing amount at the previous time and the current browsing amount and a preset difference value, and determining the reason why the recommendation result does not meet the standard based on a comparison result of the absolute value and a pre-stored preset absolute value, wherein: if the absolute value is less than or equal to the first preset absolute value, it is determined that the course package itself has a problem, and a processing notification is sent; if the absolute value is greater than the first preset absolute value and less than a second preset absolute value, the reason why the recommendation result does not meet the standard is determined based on a variance of the historical browsing amount; and if the absolute value is greater than or equal to the second preset absolute value, it is determined that there is a problem in the setting of the preset adaptation degree, and the preset adaptation degree is adjusted based on a difference between the preset browsing amount and the actual browsing amount.

[0019] Further, the process of determining the reason why the recommendation result does not meet the standard based on the variance of the historical browsing amount comprises: comparing the variance with a preset variance; if the variance is greater than the preset variance, a historical time-historical browsing amount curve for the course package is drawn, and the reason why the recommendation result does not meet the standard is determined based on a characteristic of the historical time-historical browsing amount curve; and if the variance is less than or equal to the preset variance, it is determined that the course package itself has a problem, and a processing notification is sent.

[0020] Further, the process of determining the characteristic of the curve based on the autocorrelation function comprises: determining a lag number; calculating an autocorrelation value using a normalization method according to the lag number, and drawing a lag number-autocorrelation value curve; counting a number of peaks of the curve, and comparing the number of peaks with the lag number; if the number is greater than the lag number, it is determined that the curve has periodicity; and if the number is less than or equal to the lag number, it is determined that the curve does not have periodicity.

[0021] Further, the process of determining the reason why the recommendation result does not meet the standard according to the characteristics of the lag number-autocorrelation value curve comprises: determining the reason why the recommendation result does not meet the standard based on whether the curve has periodicity; if the lag number-autocorrelation value curve has periodicity, determining that the reason is that the course package is currently in a browsing off-season, and continuously monitoring the browsing volume of the corresponding course package; if the lag number-autocorrelation value curve does not have periodicity, determining the reason according to the comparison result of the browsing volume at the last moment and the current browsing volume; if the browsing volume at the last moment is greater than the current browsing volume, determining that the reason is that the course package itself has a problem, and issuing a processing notification; if the browsing volume at the last moment is less than or equal to the current browsing volume, determining that the reason is that the course package is currently in a browsing off-season, and continuously monitoring the browsing volume of the corresponding course package.

[0022] Further, the process of adjusting the preset adaptation degree based on the difference between the corresponding preset browsing volume and actual browsing volume comprises: increasing the preset adaptation degree based on the difference, and the increase amplitude of the preset adaptation degree is proportional to the difference.

[0023] Compared with the prior art, the present application has the beneficial effects that, by preprocessing the obtained user information and online course information, the present application can improve the use value of the obtained data, and by inputting the preprocessed user information and online course information into the knowledge graph, the knowledge graph can better label these data, and based on the labels, the adaptation degree of the user and the course package is calculated, and according to the adaptation degree, the user is recommended the course package, which can make the online course recommendation more accurate, and according to the browsing volume of each recommended online course, it is determined whether the recommendation result meets the standard, and the reason why the recommendation result does not meet the standard is preliminarily determined, so that the online course recommendation can be more accurately pushed according to the user's demand, the efficiency of the user in the course learning process is improved, and the efficiency of the online course recommendation is further improved.

[0024] Further, the present application determines whether the recommendation result meets the standard by the actual browsing volume and the pre-stored preset browsing volume, and preliminarily determines the reason why the recommendation result does not meet the standard, which can quickly determine whether the online course recommendation can achieve the effect, thereby further improving the efficiency of online course recommendation.

[0025] Further, the present application determines whether the recommendation result meets the standard by the actual browsing volume and the pre-stored preset browsing volume, and preliminarily determines the reason why the recommendation result does not meet the standard, which can quickly determine whether the online course recommendation can achieve the effect, thereby further improving the efficiency of online course recommendation.

[0026] Further, the preset browse amount is adjusted by the difference between the actual use ratio and the preset use ratio, the preset browse amount can be more accurately adjusted, so that the determination result of whether the recommendation result meets the standard is more accurate, and the efficiency of online course recommendation is further improved.

[0027] Further, the preset browse amount is adjusted by the difference between the actual use ratio and the preset use ratio, the preset browse amount can be more accurately adjusted, so that the determination result of whether the recommendation result meets the standard is more accurate, and the efficiency of online course recommendation is further improved.

[0028] Further, the preset browse amount is adjusted by the difference between the actual use ratio and the preset use ratio, the preset browse amount can be more accurately adjusted, so that the determination result of whether the recommendation result meets the standard is more accurate, and the efficiency of online course recommendation is further improved.

[0029] Further, the preset browse amount is adjusted by the difference between the actual use ratio and the preset use ratio, the preset browse amount can be more accurately adjusted, so that the determination result of whether the recommendation result meets the standard is more accurate, and the efficiency of online course recommendation is further improved.

[0030] Further, the preset browse amount is adjusted by the difference between the actual use ratio and the preset use ratio, the preset browse amount can be more accurately adjusted, so that the determination result of whether the recommendation result meets the standard is more accurate, and the efficiency of online course recommendation is further improved.

[0031] Further, the preset browse amount is adjusted by the difference between the actual use ratio and the preset use ratio, the preset browse amount can be more accurately adjusted, so that the determination result of whether the recommendation result meets the standard is more accurate, and the efficiency of online course recommendation is further improved.

[0032] Further, the preset browse amount is adjusted by the difference between the actual use ratio and the preset use ratio, the preset browse amount can be more accurately adjusted, so that the determination result of whether the recommendation result meets the standard is more accurate, and the efficiency of online course recommendation is further improved. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 The step flow chart of the online course learning recommendation method based on the knowledge graph of the embodiment of the present application is shown in the figure.

[0034] Figure 2A step flow chart for determining by the embodiment of the present application based on a comparison result of a browsing volume of each of the course packages and a preset browsing volume;

[0035] Figure 3 A step flow chart for determining by the embodiment of the present application based on a cumulative learning duration of the user for the recommended course package and a preset learning duration;

[0036] Figure 4 A step flow chart for determining by the embodiment of the present application based on a difference between a previous browsing volume and a current browsing volume and a preset difference value and a preset absolute value stored in advance. DETAILED DESCRIPTION

[0037] In order to make the objects and advantages of the present application clearer, the present application will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0038] The preferred embodiments of the present application will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and do not limit the protection scope of the present application.

[0039] Please refer to Figure 1 The step flow chart of the online course learning recommendation method based on the knowledge graph of the embodiment of the present application is shown.

[0040] The steps of the embodiment of the present application in actual operation process include:

[0041] S1, obtaining user information and online course information;

[0042] S2, respectively pre-processing the user information and the online course information;

[0043] S3, inputting the pre-processed user information and the online course information into the knowledge graph, the knowledge graph packaging the online courses into a plurality of course packages according to the characteristics of the online courses and outputting a plurality of preset labels for each of the course packages and a plurality of demand labels for the user according to the user information, wherein the preset labels are common characteristics of all courses in the course package;

[0044] S4, calculating an adaptation degree of the course package and the user demand, wherein the adaptation degree is a ratio of a number of same labels in the preset labels of the course package and the demand labels of the user to a total number of the demand labels of the user;

[0045] S5, recommending the course package to the user according to the adaptation degree, and pushing the course package to the user when the adaptation degree is higher than a preset adaptation degree;

[0046] S6, determining whether the recommendation result of the online course meets the standard based on the browsing volume of each recommended course package, and determining the corresponding processing mode based on the reason for not meeting the standard;

[0047] S7, adjusting the online course learning recommendation method based on the processing mode.

[0048] Specifically, first, user information and online course information are obtained through a website analysis tool, and then the user information is preprocessed, including the following steps: first, data cleaning is used to ensure that the information of each user is unique, and missing values are processed by filling them with the mean; second, data transformation is used to normalize numerical data (such as age) to eliminate scale differences between different features, and to encode categorical data (such as gender and occupation); then, according to the course content and data characteristics, features useful for user behavior analysis are selected; finally, the cleaned and transformed user data is integrated together and stored. The online course information is preprocessed, including the following steps: first, use data cleaning methods to remove duplicate course information, ensure that the information of each course is unique, and for missing course information (such as teacher name and course description), adopt filling or deletion strategy for processing; second, use data transformation for numerical data (such as class hours and credits) for normalization, and for text data (such as course description) for word segmentation and word vectorization, etc. for text analysis.

[0049] Specifically, the preprocessed user information and online course information are input into the constructed knowledge graph. The system first analyzes the online course information in depth. Each course is considered as a node in the graph, and the relationships between courses such as association, dependence, and similarity are represented by edges. The system uses clustering algorithms or topic models and other advanced technologies to intelligently package these courses into several course packages with internal logical connections according to the characteristics of the courses, such as subject area, teaching method, target audience, etc. Each course package is assigned a series of preset labels that accurately reflect the common characteristics of all courses in the package. For example, a course package may contain all the basic courses on "Data Analysis and Machine Learning", so its preset labels may include "Computer Science", "Data Analysis", "Machine Learning Introduction", etc. At the same time, the user information is deeply mined according to the user's personal characteristics, historical behavior, interests and hobbies, etc. to generate several demand labels for the user. These demand labels accurately describe the user's current learning needs, interest preferences, and potential learning goals. For example, a user who is interested in artificial intelligence may have demand labels such as "AI technology frontier", "Python programming", "deep learning", etc.

[0050] Please refer toFigure 2 As shown, it is a step flow chart of the embodiment of the application based on the comparison result of the browsing volume of each course package and the preset browsing volume. The process of determining whether the recommendation result of the online course meets the standard based on the browsing volume of each course package includes: determining whether the recommendation result meets the standard based on the comparison result of the browsing volume and the pre-stored preset browsing volume; if the browsing volume is greater than or equal to the first preset browsing volume, it is determined that the recommendation result meets the standard; if the browsing volume is less than the first preset browsing volume and greater than the second preset browsing volume, it is determined whether the recommendation result meets the standard based on the cumulative learning time of the user for the recommended course package, wherein the cumulative learning time is the average time of the user learning the course; if the browsing volume is less than or equal to the second preset browsing volume, it is determined that the recommendation result does not meet the standard, and the reason why the recommendation result does not meet the standard is determined based on the difference between the browsing volume at the last moment and the current browsing volume.

[0051] Specifically, in the embodiment, the browsing volume L0 can be divided into the first preset browsing volume L1 and the second preset browsing volume L2, the first preset browsing volume L1 is set to 3000 in the browsing volume standard, and the second preset browsing volume L2 is set to 0.7xL1. It should be noted that in other embodiments, the values of L1 and L2 can also be determined according to the relative demand of online courses, and the numerical values of the browsing volume are all rounded up; the comparison process based on the browsing volume L and L1 and L2 is as follows:

[0052] If the browsing volume L is greater than or equal to the first preset browsing volume L1, it is determined that the recommendation result meets the standard based on the browsing volume;

[0053] If the browsing volume L is less than the first preset browsing volume L1 and greater than the second preset browsing volume L2, it is necessary to determine whether the recommendation result meets the standard based on the cumulative learning time P of the user for the recommended course package, wherein the cumulative learning time is the average time of the user learning the course;

[0054] If the browsing volume L is less than or equal to the second preset browsing volume L2, it is determined that the recommendation result does not meet the standard based on the browsing volume, and the reason why the recommendation result does not meet the standard is determined based on the difference Q between the browsing volume at the last moment and the current browsing volume.

[0055] Please refer to Figure 3As shown, it is the step flow chart of the embodiment of the application based on the accumulated learning time of the user for the recommended course package and the preset learning time. The process of the embodiment of the application based on the accumulated learning time of the user for the recommended course package to determine whether the recommended result meets the standard includes: comparing the accumulated learning time with the preset learning time; if the accumulated learning time is less than or equal to the preset learning time, it is determined that the recommended result does not meet the standard, and the reason why the recommended result does not meet the standard is determined based on the difference between the last browsing volume and the current browsing volume; if the accumulated learning time is greater than the preset learning time, it is determined that the recommended result meets the standard, and the preset browsing volume is adjusted based on the actual use ratio, wherein the actual use ratio is the ratio of the number of users whose accumulated learning time is greater than the preset learning time to the total number of users who receive the course package push.

[0056] Specifically, in the case where the browsing volume L is less than the first preset browsing volume L1 and greater than the second preset browsing volume L2, it is determined whether the recommended result meets the standard based on the accumulated learning time and the preset learning time; in this embodiment, the preset learning time P0 is set to 5h, and the comparison process based on the accumulated learning time and the preset learning time is specifically as follows:

[0057] If the accumulated learning time P is less than or equal to the preset learning time P0, it indicates that the learning time of the user does not meet the standard, and it is determined that the recommended result does not meet the standard, and the reason why the recommended result does not meet the standard is determined based on the difference R between the last browsing volume and the current browsing volume;

[0058] If the accumulated learning time P is greater than the preset learning time P0, it indicates that the learning time of the user meets the standard, and it is determined that the recommended result meets the standard, and the preset browsing volume is adjusted based on the actual use ratio.

[0059] Specifically, the process of the embodiment of the application based on the actual use ratio to adjust the preset browsing volume includes: calculating the difference between the actual use ratio and the preset use ratio; and reducing the preset browsing volume based on the obtained difference, and the reduction amplitude of the difference and the preset browsing volume is proportional.

[0060] Specifically, in this embodiment, the preset difference T0 is 0.2, and the comparison process based on the difference between the actual use ratio and the preset use ratio and the preset difference is specifically as follows:

[0061] If the difference T is less than or equal to the preset difference T0, the preset browsing volume is adjusted to 0.8 times the original preset browsing volume;

[0062] If the difference T is greater than the preset difference T0, the preset browsing volume is adjusted to 0.5 times the original preset browsing volume.

[0063] Specifically, the process of adjusting the preset browsing amount in the embodiment of the present application further includes: calculating the difference between the average interaction ratio of the user and the online course and the preset interaction ratio; and reducing the preset browsing amount based on the obtained difference, and the difference is directly proportional to the reduction amplitude of the browsing amount.

[0064] Specifically, in the embodiment, the interaction ratio is the completion degree of the interaction content in the course, that is, the ratio of the completed interaction content to the total number of interactions in the course, wherein the interaction content is the questions and answers and the in-class tests in the course teaching process, the preset difference U0 between the average interaction ratio and the preset interaction ratio is set to 0.2, and the process based on the comparison between the difference and the preset difference is specifically as follows:

[0065] If the difference U is less than or equal to the preset difference U0, the preset browsing amount is adjusted to 0.76 times the original preset browsing amount.

[0066] If the difference U is greater than the preset difference U0, the preset browsing amount is adjusted to 0.57 times the original preset browsing amount.

[0067] Please refer to Figure 4 The process of determining the reason why the recommended result does not meet the standard when the browsing amount is less than or equal to the second preset browsing amount in the embodiment of the present application includes: calculating the absolute value of the difference between the browsing amount at the last time and the current browsing amount and the preset difference, and determining the reason why the recommended result does not meet the standard based on the comparison result between the absolute value and the pre-stored preset absolute value, wherein: if the absolute value is less than or equal to the first preset absolute value, it is determined that there is a problem with the course package itself, and a processing notification is sent; if the absolute value is greater than the first preset absolute value and less than the second preset absolute value, the reason why the recommended result does not meet the standard is determined based on the variance of the historical browsing amount; and if the absolute value is greater than or equal to the second preset absolute value, it is determined that there is a problem with the setting of the preset adaptation degree, and the preset adaptation degree is adjusted based on the difference between the corresponding preset browsing amount and the actual browsing amount.

[0068] Specifically, in the embodiment, the absolute value W0 can be divided into a first preset absolute value W1 and a second preset absolute value W2, the first preset absolute value W1 in the absolute value standard is set to 100, and the second preset absolute value W2 is set to 1.5*W1, and it should be noted that in other embodiments, the values of W1 and W2 can also be determined according to the needs of the online course; the comparison process between the absolute value and the pre-stored preset absolute value is specifically as follows:

[0069] If the absolute value W is less than or equal to the first preset absolute value W1, it is indicated that the browsing quantity at the last time and the current time is low and stable, and it is determined that the course package itself has a problem, and a processing notification is sent;

[0070] If the absolute value W is greater than the first preset absolute value W1 and less than a second preset absolute value W2, it is indicated that the case cannot determine the reason why the recommended result does not meet the standard, and the reason why the recommended result does not meet the standard needs to be determined again based on the variance G of the historical browsing quantity;

[0071] If the absolute value W is greater than or equal to the second preset absolute value W2, it is determined that the setting of the preset adaptation degree has a problem, and the preset adaptation degree is adjusted based on the difference H between the preset browsing quantity and the actual browsing quantity.

[0072] Specifically, the process of determining the reason why the recommended result does not meet the standard based on the variance of the historical browsing quantity includes: comparing the variance with a preset variance; if the variance is greater than the preset variance, drawing a historical time-historical browsing quantity curve for the data package, and determining the reason why the recommended result does not meet the standard based on the characteristics of the historical time-historical browsing quantity curve; and if the variance is less than or equal to the preset variance, determining that the course package itself has a problem, and sending a processing notification.

[0073] Specifically, in the embodiment, the preset variance G0=0.94, and the comparison process based on the variance and the preset variance is specifically as follows:

[0074] If the variance G is greater than the preset variance G0, it is indicated that the numerical dispersion of the historical browsing quantity is high, the historical time-historical browsing quantity curve is drawn, and the reason why the recommended result does not meet the standard is determined based on the characteristics of the historical time-historical browsing quantity curve;

[0075] If the variance G is less than or equal to the preset variance G0, it is indicated that the numerical dispersion of the historical browsing quantity is low, it is determined that the course package itself has a problem, and the problem may be that the online course in the course package has a problem, and a processing notification for reorganizing the courses in the course package is sent.

[0076] Specifically, the process of determining the characteristics of the curve based on the autocorrelation function includes: determining a lag number; calculating an autocorrelation value using a normalization method according to the lag number, and drawing a lag number-autocorrelation value curve; counting the number of peaks of the curve, and comparing the number of peaks with the lag number; if the number is greater than the lag number, it is determined that the curve has periodicity; and if the number is less than or equal to the lag number, it is determined that the curve does not have periodicity.

[0077] Specifically, the autocorrelation function is a cross-correlation of a signal (or time series) with itself at different points in time. It measures the similarity between a signal at two different points in time as a function of the time difference between the two points in time. In statistics, autocorrelation is defined as the Pearson correlation between values of a random process at different times.

[0078] Specifically, the process of determining the reason why the recommendation result does not meet the standard according to the characteristics of the lag number-autocorrelation value curve in the embodiment of the present application includes: determining the reason why the recommendation result does not meet the standard based on whether the curve has periodicity; if the lag number-autocorrelation value curve has periodicity, determining that the reason is that the course package is currently in a browsing off-season, and then continuously monitoring the browsing volume of the corresponding course package; if the lag number-autocorrelation value curve does not have periodicity, determining the reason according to the comparison result of the browsing volume at the last moment and the current browsing volume; if the browsing volume at the last moment is greater than the current browsing volume, determining that there is a problem with the course package itself, and issuing a processing notification; if the browsing volume at the last moment is less than or equal to the current browsing volume, determining that the reason is that the course package is currently in a browsing off-season, and continuously monitoring the browsing volume of the corresponding course package.

[0079] Specifically, in the embodiment, the determination process based on whether the lag number-autocorrelation value curve has periodicity is as follows:

[0080] If the lag number-autocorrelation value curve has periodicity, it is determined that the course package is currently in a user usage off-season, and then the browsing volume of the data package is continuously monitored subsequently;

[0081] If the lag number-autocorrelation value curve does not have periodicity, the browsing volume at the last moment is compared with the current browsing volume, wherein:

[0082] If the browsing volume at the last moment is greater than the current browsing volume, it indicates that the browsing volume is in a downward trend, and it is determined that there is a problem with the course package itself, which leads to this situation, and a processing notification of reorganizing the data in the data package is issued;

[0083] If the browsing volume at the last moment is less than or equal to the current browsing volume, it indicates that the browsing volume of the course package by the user is in a fluctuation situation, and the browsing volume of the data package is continuously monitored subsequently.

[0084] Specifically, the process of adjusting the preset adaptation degree based on the difference between the corresponding preset browsing volume and the actual browsing volume in the embodiment of the present application includes: increasing the preset adaptation degree based on the difference, and the increase amplitude of the difference and the preset adaptation degree is proportional.

[0085] Specifically, in the embodiment, the preset difference H0 is 500, and the comparison between the difference H and the preset difference H0 is specifically as follows:

[0086] If the difference H is less than or equal to the preset difference H0, the preset adaptation degree is adjusted to 1.2 times of the original preset adaptation degree.

[0087] If the difference H is greater than the preset difference H0, the preset adaptation degree is adjusted to 1.7 times of the original preset adaptation degree.

[0088] In order to make the purpose and advantages of the present application more clear and apparent, the present application will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0089] Embodiment 1

[0090] The embodiment is directed to the calculation of the adaptation degree of a user and a course package, wherein the demand tags of the user are as follows: deep learning, Python programming, AI technology frontier, object detection, and object tracking; and the preset tags of the course package are as follows: computer science, data analysis, object detection, Python programming, and AI technology frontier. At this time, the demand tags of the user and the preset tags of the course package are compared, wherein the same tags are Python programming, AI technology frontier, and object detection. Therefore, the adaptation degree of the user and the course package is the same tags of the user and the course package / the total number of the demand tags of the user, i.e. 3 / 5 = 0.6.

[0091] So far, the technical solutions of the present application have been described in conjunction with the preferred embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical solutions after the changes or replacements will fall within the protection scope of the present application.

[0092] The above is only the preferred embodiments of the present application and is not used to limit the present application; for those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1.A knowledge graph-based online course learning recommendation method, characterized in that, The method comprises the following steps: acquiring user information and online course information; preprocessing the user information and the online course information respectively; inputting the preprocessed user information and online course information into a knowledge graph, the knowledge graph packaging the online course information into a plurality of course packages according to the characteristics of the online course information, outputting a plurality of preset labels for each course package and a plurality of demand labels for the user according to the user information, wherein the preset labels are common characteristics of all courses in the course package; calculating the adaptation degree of the course package and the user demand, wherein the adaptation degree is the ratio of the number of identical labels in the preset labels of the course package and the demand labels of the user to the total number of the demand labels; recommending the course package to the user according to the adaptation degree, and pushing the course package to the user when the adaptation degree is higher than a preset adaptation degree; judging whether the recommendation result of the online course meets the standard based on the browsing volume of each recommended course package, and determining the corresponding processing mode based on the reason why the recommendation result does not meet the standard; adjusting the online course learning recommendation method based on the processing mode; the process of judging whether the recommendation result of the online course meets the standard based on the browsing volume of each course package comprises: judging whether the recommendation result meets the standard based on the comparison result of the browsing volume and a preset browsing volume; if the browsing volume is greater than or equal to a first preset browsing volume, it is determined that the recommendation result meets the standard; if the browsing volume is less than the first preset browsing volume and greater than a second preset browsing volume, then judging whether the recommendation result meets the standard based on the cumulative learning time of the user for the recommended course package, wherein the cumulative learning time is the average learning time of the user for the course; if the browsing volume is less than or equal to the second preset browsing volume, it is determined that the recommendation result does not meet the standard, and the reason why the recommendation result does not meet the standard is determined based on the difference between the browsing volume at the last time and the current browsing volume; the process of judging whether the recommendation result meets the standard based on the cumulative learning time of the user for the recommended course package comprises: comparing the cumulative learning time with a preset learning time; if the cumulative learning time is less than or equal to the preset learning time, it is determined that the recommendation result does not meet the standard, and the reason why the recommendation result does not meet the standard is determined based on the difference between the browsing volume at the last time and the current browsing volume; if the cumulative learning time is greater than the preset learning time, it is determined that the recommendation result meets the standard, and the preset browsing volume is adjusted based on the actual usage ratio, wherein the actual usage ratio is the ratio of the number of users whose cumulative learning time is greater than the preset learning time to the total number of users who have received the course package push; the process of determining the reason why the recommendation result does not meet the standard when the browsing volume is less than or equal to the second preset browsing volume comprises: calculating the absolute value of the difference between the browsing volume at the last time and the current browsing volume and a preset difference value, and determining the reason why the recommendation result does not meet the standard based on the comparison result of the absolute value and a preset absolute value, wherein: if the absolute value is less than or equal to the first preset absolute value, it is determined that there is a problem with the course package itself, and a processing notification is sent. If the absolute value is greater than the first preset absolute value and less than a second preset absolute value, a reason for the recommendation result not meeting the standard is determined based on a variance of historical browsing amounts; If the absolute value is greater than or equal to the second preset absolute value, it is determined that there is a problem with the setting of the preset adaptation degree, and the preset adaptation degree is adjusted based on a difference between the preset browsing amount and the actual browsing amount. 2.The knowledge graph-based online course learning recommendation method according to claim 1, characterized in that, The process of adjusting the preset browsing amount based on the actual usage ratio includes: calculating a difference between the actual usage ratio and the preset usage ratio; reducing the preset browsing amount based on the obtained difference, and the difference is directly proportional to the reduction amplitude of the preset browsing amount. 3.The knowledge graph-based online course learning recommendation method of claim 2, characterized in that, The process of adjusting the preset browsing amount also includes: calculating a difference between the average interaction ratio of the user and the online course and the preset interaction ratio; reducing the preset browsing amount based on the obtained difference, and the difference is directly proportional to the reduction amplitude of the browsing amount. 4.The knowledge graph-based online course learning recommendation method of claim 1, wherein, The process of determining the reason for the recommendation result not meeting the standard based on the variance of the historical browsing amounts includes: comparing the variance with a preset variance; if the variance is greater than the preset variance, a historical time-historical browsing amount curve for the course package is drawn, and a reason for the recommendation result not meeting the standard is determined based on the characteristics of the historical time-historical browsing amount curve; if the variance is less than or equal to the preset variance, it is determined that there is a problem with the course package itself, and a processing notification is issued. 5.The knowledge graph-based online course learning recommendation method according to claim 4, characterized in that, The process of determining the characteristics of the curve based on the autocorrelation function includes: determining a lag number; according to the lag number, calculating an autocorrelation value using a normalization method, and drawing a lag number-autocorrelation value curve; counting the number of peaks of the curve, and comparing the number of peaks with the lag number; if the number is greater than the lag number, it is determined that the curve has periodicity; if the number is less than or equal to the lag number, it is determined that the curve does not have periodicity. 6.The knowledge graph-based online course learning recommendation method according to claim 5, characterized in that, The process of determining the reason for the recommendation result not meeting the standard based on the characteristics of the lag number-autocorrelation value curve includes: determining the reason for the recommendation result not meeting the standard based on whether the curve has periodicity; if the lag number-autocorrelation value curve has periodicity, it is determined that the reason is that the course package is currently in a browsing off-season, and the browsing amount of the corresponding course package is continuously monitored; if the lag number-autocorrelation value curve does not have periodicity, the reason is determined according to the comparison result of the last time browsing amount and the current browsing amount; if the last time browsing amount is greater than the current time browsing amount, it is determined that the reason is that there is a problem with the course package itself, and a processing notification is issued; if the last time browsing amount is less than or equal to the current time browsing amount, it is determined that the reason is that the course package is currently in a browsing off-season, and the browsing amount of the corresponding course package is continuously monitored. 7.The knowledge graph-based online course learning recommendation method according to claim 1, characterized in that, The process of adjusting the preset adaptation degree based on the difference between the preset browsing amount and the actual browsing amount includes: increasing the preset adaptation degree based on the difference, and the difference is directly proportional to the increase amplitude of the preset adaptation degree.

Images