Online course learning behavior determination method and device

By obtaining and analyzing course learning behavior data on the MOOC platform, determining the learning behavior of target students, and using clustering algorithms and dynamic time regularization algorithms, the problem of failure to effectively analyze learning behavior in the existing technology is solved, and the problem of each The learning behavior analysis and auxiliary learning guidance of online courses have improved students' learning effectiveness.

CN114510617BInactive Publication Date: 2024-11-01HARBIN INST OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210098102.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-27
Publication Date
2024-11-01
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing technology has failed to effectively analyze and utilize course learning behavior data on the MOOC platform, and lacks learning behavior analysis and auxiliary learning guidance for each online course.

Method used

By obtaining the historical data of the target online course, determining the target students who can refer to the learning behavior, calculating the relative distances of any two students when learning, using dynamic time regularization algorithm to cluster learning behaviors, obtaining multiple types of learning behaviors, and Determine appropriate learning behaviors based on these behaviors and finally provide students with learning tips.

Benefits of technology

It realizes learning behavior analysis for each online course, provides auxiliary learning guidance, and improves students' learning efficiency and grades.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114510617B_ABST
    Figure CN114510617B_ABST
Patent Text Reader

Abstract

The present invention provides a method and device for determining learning behaviors of online courses, wherein the method includes: obtaining historical data related to a target online course; determining target students whose learning behaviors for the target online course can be referenced among students who are learning the target online course according to the historical data; determining the relative distance between any two target students when learning the target online course; clustering the learning behaviors of the target online course according to the relative distance between any two target students when learning the target online course, and obtaining multiple types of learning behaviors corresponding to the target online course; determining the learning behaviors suitable for the target online course according to the obtained multiple types of learning behaviors. This scheme can determine suitable learning behaviors for each target online course, and then can use the determined learning behaviors to prompt students who are learning the corresponding target online course, so as to assist students in learning.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data processing technology, and in particular to a method and device for determining online course learning behavior. Background Art

[0002] With the innovation of information technology, Massive Open Online Courses (MOOC) platforms have gradually become a popular teaching model. Its advantage is that online education resources are freer and easier to obtain than traditional physical education. More and more people are starting to use MOOC platforms for teaching and learning. It cannot be ignored that MOOC is not only a product of the development of online education, but also a product of the big data era. The implementation of MOOC depends on the network platform, and it is precisely this feature that makes it easy to collect data. Compared with traditional education, its data richness and data volume have obvious advantages. With the development and deep integration of data technology and MOOC platforms, data management and analysis technologies have been applied to various parts of the MOOC platform, including course content, homework, after-class discussions, Q&A exchanges, and other aspects, which contain a large amount of behavioral data of teachers and students.

[0003] At present, most of the research focuses on using these data to predict students' final grades, and there is no analysis of course learning behavior. The present invention provides a method for determining online course learning behavior. Summary of the invention

[0004] The embodiment of the present invention provides a method and device for determining online course learning behavior, which can determine appropriate learning behavior for each online course and play a role in assisting students' learning.

[0005] In a first aspect, an embodiment of the present invention provides a method for determining online course learning behavior, comprising:

[0006] Acquire historical data related to the target online course; the historical data at least includes information about students who study the target online course and behavior information about the students in the process of studying the target online course;

[0007] determining, according to the historical data, among students studying the target online course, target students whose learning behaviors for the target online course can be used as references;

[0008] Determine the relative distance between any two target students when they are learning the target online course according to the behavior information of each target student in the process of learning the target online course;

[0009] Clustering the learning behaviors of learning the target online course according to the relative distance between any two target students when learning the target online course, to obtain multiple types of learning behaviors corresponding to learning the target online course;

[0010] According to the obtained multiple types of learning behaviors, a learning behavior suitable for the target online course is determined.

[0011] Preferably, the step of determining a target student whose learning behavior for the target online course can be referenced comprises:

[0012] If it is determined that the student's test score for the target online course meets a preset level, and / or it is determined that the student's learning activity for the target online course meets a preset activity, the student is determined as the target student.

[0013] Preferably, determining the relative distance between any two target students when studying the target online course includes:

[0014] Determine the individual chapter videos included in the target online course;

[0015] Abstracting the behavior information of each target student learning each chapter video into discrete scattered points, and generating corresponding time series according to the abstracted scattered points;

[0016] According to the time sequence corresponding to each target student learning each chapter video, the time sequences corresponding to any two target students learning the same chapter video are aligned using a dynamic time warping algorithm to obtain aligned target scatter points when any two target students learn the same chapter video;

[0017] According to the aligned target scatter points when any two target students study the same chapter video, the relative distance between any two target students when they study the target online course is calculated.

[0018] Preferably, the step of abstracting the behavior information of each target student learning each chapter video into discrete scattered points, and generating a corresponding time series according to the abstracted scattered points includes:

[0019] According to the behavior information of each target student on the chapter video, the number of views of each target student on the chapter video at each viewing time point is determined; the viewing time point is not less than 0 and not greater than the maximum playback time point of the chapter video;

[0020] The viewing time point and the corresponding viewing number are abstracted into a scattered point, and then for each of the target students, a time series of learning the video of this chapter is obtained.

[0021] Preferably, the method of using a dynamic time warping algorithm to align the time series corresponding to any two target students learning the same chapter video to obtain the aligned target scatter points when any two target students learn the same chapter video includes:

[0022] Construct an n×m matrix grid for the time series corresponding to any two target students when they learn the same chapter video; wherein n and m are the number of scattered points contained in the time series corresponding to the two target students when they learn the same chapter video; the (i, j)th matrix element in the matrix grid is the distance between the i-th scattered point in the time series corresponding to one of the target students and the j-th scattered point in the time series corresponding to the other target student;

[0023] For the matrix grid, determining the shortest path from the (1, 1)th matrix element to the (n, m)th matrix element through the matrix elements;

[0024] The scattered points of the two learning sequences corresponding to each matrix element included in the shortest path are determined as the aligned target scattered points.

[0025] Preferably, the calculating of the relative distance between any two target students when studying the target online course includes:

[0026] The relative distance between any two target students when learning the target online course is calculated using the following formula:

[0027]

[0028] Wherein, D(A, B) is the relative distance between target student A and target student B when they study the target online course, taking the minimum value in the function F[·]; P is the number of chapter videos contained in the target online course; w(p) is the aligned target scatter point when target student A and target student B study the pth chapter video; a(p) and b(p) are the viewing time of target student A and target student B when they study the pth chapter video respectively; d(a(p), b(p)) is the difference in viewing time when target student A and target student B study the pth chapter video.

[0029] Preferably, clustering the learning behaviors of learning the target online course according to the relative distance between any two target students when learning the target online course to obtain multiple types of learning behaviors corresponding to learning the target online course includes:

[0030] Determine the adaptive K value;

[0031] Take each target student as the center, sort the relative distances between other target students and the target student at the center by size, and select the other target students corresponding to the first K smallest relative distances and the target student at the center as one category;

[0032] Determine the classification results of all target students according to the categories with the highest number of occurrences, and obtain the learning behaviors of the target students included in each category in learning the target online course;

[0033] and / or,

[0034] Determining a learning behavior suitable for the target online course based on the obtained multiple types of learning behaviors includes:

[0035] According to each type of learning behavior, determining the number of views corresponding to each viewing time point of the target online course, and determining the number of views corresponding to each viewing time point of the target online course as a learning behavior suitable for the target online course;

[0036] and / or,

[0037] After determining the learning behavior suitable for the target online course, the method further includes: providing prompts to students studying the target online course according to the learning behavior suitable for the target online course.

[0038] In a second aspect, an embodiment of the present invention further provides an online course learning behavior determination device, comprising:

[0039] A historical data acquisition unit, used to acquire historical data related to a target online course; the historical data at least includes information about students who study the target online course and behavioral information about students who study the target online course;

[0040] a target student determination unit, configured to determine, based on the historical data, a target student whose learning behavior for the target online course can be used as a reference among students studying the target online course;

[0041] A relative distance determination unit, configured to determine the relative distance between any two target students when they are learning the target online course according to the behavior information of each target student in the process of learning the target online course;

[0042] A clustering unit, configured to cluster the learning behaviors of learning the target online course according to the relative distance between any two target students when learning the target online course, and obtain a plurality of types of learning behaviors corresponding to learning the target online course;

[0043] The learning behavior determination unit is used to determine a learning behavior suitable for the target online course according to the obtained multiple types of learning behaviors.

[0044] In a third aspect, an embodiment of the present invention further provides a computing device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method described in any embodiment of this specification is implemented.

[0045] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, enables the computer to execute the method described in any embodiment of this specification.

[0046] The embodiment of the present invention provides a method and device for determining learning behavior of an online course. After determining the target students whose learning behavior for the target online course can be referenced, the learning behavior of these target students for the target online course can be clustered. However, the learning behavior is behavior information with time series characteristics. Therefore, it is necessary to determine the relative distance between any two target students when learning the target online course, and cluster them using the relative distance, so that multiple types of learning behaviors corresponding to learning the target online course can be obtained. Finally, based on the multiple types of learning behaviors, the learning behavior suitable for the target online course is determined. It can be seen that this scheme can determine the appropriate learning behavior for each target online course, and then can use the determined learning behavior to prompt the students who are learning the corresponding target online course, so as to play a role in assisting students' learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 This is a flow chart of a method for determining online course learning behavior provided by an embodiment of the present invention;

[0049] Figure 2 is a flow chart of a relative distance solution process provided by an embodiment of the present invention;

[0050] Figure 3 is a hardware architecture diagram of a computing device provided by an embodiment of the present invention;

[0051] Figure 4 It is a structural diagram of an online course learning behavior determination device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0053] Different courses have different key points and difficulties. If we can analyze the learning behavior of each course and find out the appropriate learning behavior for each course, we can provide guidance on the learning behavior of the course for subsequent students who study the course, and promote the role of auxiliary learning. In order to determine the appropriate learning behavior for each course, we can consider using multiple students who have completed the course and have excellent grades to analyze the learning behavior of the course, and determine the appropriate learning behavior through clustering and other methods.

[0054] The specific implementation of the above concept is described below.

[0055] Please refer to Figure 1 , an embodiment of the present invention provides a method for determining online course learning behavior, the method comprising:

[0056] Step 100, obtaining historical data related to a target online course; the historical data at least includes information about students who study the target online course and behavioral information about students who study the target online course;

[0057] Step 102, determining, based on the historical data, among the students studying the target online course, target students whose learning behaviors for the target online course can be used as references;

[0058] Step 104, determining the relative distance between any two target students when they are learning the target online course according to the behavior information of each target student in the process of learning the target online course;

[0059] Step 106, clustering the learning behaviors of learning the target online course according to the relative distance between any two target students when learning the target online course, to obtain multiple types of learning behaviors corresponding to learning the target online course;

[0060] Step 108: Determine a learning behavior suitable for the target online course based on the obtained multiple types of learning behaviors.

[0061] In the embodiment of the present invention, after determining the target students whose learning behaviors for the target online course can be referenced, the learning behaviors of these target students for the target online course can be clustered. However, the learning behavior is behavior information with time series characteristics. Therefore, it is necessary to determine the relative distance between any two target students when learning the target online course, and cluster them using the relative distance, so that multiple types of learning behaviors corresponding to learning the target online course can be obtained. Finally, based on the multiple types of learning behaviors, the learning behaviors suitable for the target online course can be determined. It can be seen that this solution can determine suitable learning behaviors for each target online course, and then can use the determined learning behaviors to prompt students who are learning the corresponding target online courses, so as to assist students in learning.

[0062] Described below Figure 1 How the various steps are performed.

[0063] First, step 100 of "obtaining historical data related to a target online course" and step 102 of "determining, according to the historical data, a target student whose learning behavior for the target online course can be referenced among students studying the target online course" are explained simultaneously.

[0064] Since it is necessary to obtain historical data related to the target online course and analyze the learning behavior of the target online course, the target online course is a course with historical data and behavioral information of the learning behavior. For example, the target online course is any one of the multiple online courses provided on the MOOC platform.

[0065] The historical data obtained needs to include the following: 1. The historical data at least includes the information of students who study the target online course; 2. The historical data includes the behavior information of each student in the process of studying the target online course.

[0066] When students study the target online course, they need to log in to their account, and the account is associated with the student's real identity. Therefore, the student information of the target online course can be obtained. The purpose of obtaining student information is to associate the learning behavior with the student, so as to facilitate the confirmation of the learning behavior.

[0067] In addition, in order to analyze the learning behavior of the target online course, it is also necessary to include the behavior information of the students in the learning process, which may include the chapter videos, viewing time points, the number of views at each viewing time point, the viewing time, the learning time, etc. of the target online course. Among them, the viewing time point refers to the time point in the chapter video (for example, the total length of the chapter video is 60 minutes, and the viewing time point is each time point from the 0th to the 10th minute), and the learning time refers to the natural time (for example, 08:00 on January 1, 2022).

[0068] In one embodiment of the present invention, after acquiring historical data related to the target online course, the historical data may be cleaned to delete duplicate invalid data, assign values ​​to missing data, merge related information, etc. In one implementation, the Python language may be used to perform cleaning of the historical data.

[0069] In one embodiment of the present invention, in order to determine the learning behavior suitable for the target online course, it is necessary to screen out the target students whose learning behavior for the target online course can be used as a reference from the students who are learning the target online course. Since the learning behavior can be used to guide other students to learn the target online course, the learning behavior of the target online course by students with excellent grades can be used as a reference. Specifically, the determination of the target students whose learning behavior for the target online course can be used as a reference includes: if it is determined that the student's test score for the target online course meets a preset level, and / or it is determined that the student's learning activity for the target online course meets a preset activity, then the student is determined as the target student.

[0070] The preset level can be a pre-set excellent level or score in this embodiment, for example, the preset level is not less than the product of the full score and 80%. If the full score is 100 points, students with a test score of not less than 80 points can be determined as target students.

[0071] The preset activity level may be the number of logins, the number of learning times of the target online course, the learning time of the target online course, etc., which are preset in this embodiment. For example, if a student logs in and learns the target online course every day for a set number of consecutive days, and the learning time of each day is not less than the set time, it is determined that the student's learning activity level for the target online course meets the preset activity level.

[0072] Then, for step 104, the relative distance between any two target students when learning the target online course is determined according to the behavior information of each target student in the process of learning the target online course.

[0073] According to step 100, the behavior information may include information such as the chapter videos of the target online course that the student has studied, the viewing time points, the number of views at each viewing time point, the viewing time, the learning time, etc. Therefore, the learning behavior of the target student studying the target online course belongs to time series data. If the time series data needs to be clustered, the relative distance of the time series data needs to be determined.

[0074] In one embodiment of the present invention, a dynamic time warping (DTW) algorithm may be used to align the time series data. Specifically, the determining of the relative distance between any two target students when learning the target online course includes the following steps S1 to S4:

[0075] S1: Determine the chapter videos included in the target online course.

[0076] It should be noted that an online course can consist of multiple chapter videos. For example, the target online course contains 10 chapter videos.

[0077] S2: Abstracting the behavior information of each target student in learning each chapter video into discrete scattered points, and generating corresponding time series according to the abstracted scattered points.

[0078] Specifically, this step S2 can be implemented in the following manner: according to the behavioral information of each target student on the chapter video, the number of views of each target student on the chapter video at each viewing time point is determined; the viewing time point is not less than 0 and not greater than the maximum playback time point of the chapter video; the viewing time point and the corresponding number of views are abstracted into a scattered point, and for each target student, the time series of learning the chapter video is obtained.

[0079] For example, the time series of the target student A learning the first chapter video includes the following scattered points: (00:01, 1 time), (00:02, 1 time), (00:03, 1 time), (00:04, 2 times), (00:05, 2 times), (00:06, 1 time)...

[0080] By abstracting the behavior information into scattered points, it is convenient to calculate the relative distance and improve the accuracy of clustering.

[0081] It should be noted that in addition to the above implementation method, other methods can also be used, such as using piecewise linear representation for fitting, using line segments to fit these discrete scattered points, specifically, the horizontal axis represents the chapter video of the target online course, and the vertical axis represents the viewing time of the target student learning the chapter video. Then each target student can get a scattered point fitted by a line segment.

[0082] S3: According to the time series corresponding to each target student when learning each chapter video, the dynamic time warping algorithm is used to align the time series corresponding to any two target students when learning the same chapter video, so as to obtain the aligned target scatter points when any two target students learn the same chapter video.

[0083] Among them, the dynamic time warping algorithm is used to measure the similarity of two time series of different lengths. The unknown quantity is stretched or shortened (compressed) until it matches the length of the reference template. In this process, the unknown sequence will be distorted or bent so that its feature quantity corresponds to the standard pattern.

[0084] In one embodiment of the present invention, when the dynamic time warping algorithm is used for alignment processing, this step S3 specifically includes: constructing an n×m matrix grid for the time series corresponding to any two target students when learning the same chapter video; wherein n and m are the number of scattered points contained in the time series corresponding to the two target students when learning the same chapter video; the (i, j)th matrix element in the matrix grid is the distance between the i-th scattered point in the time series corresponding to one of the target students and the j-th scattered point in the time series corresponding to the other target student; for the matrix grid, determining the shortest path through the matrix elements from the (1, 1)th matrix element to the (n, m)th matrix element; and determining the scattered points of the two learning sequences corresponding to each matrix element contained in the shortest path as the aligned target scattered points.

[0085] For example, target student A and target student B respectively target time series A and time series B of the first chapter video, time series A contains n scattered points, time series B contains m scattered points, and n and m are both integers greater than 0. The (i, j)th matrix element in the matrix grid is the distance between the i-th scattered point in time series A and the j-th scattered point in time series B. The distance is used to characterize the similarity between the two scattered points. The smaller the distance, the higher the similarity. In one implementation, the distance is calculated using the Euclidean distance, which is calculated using the following formula:

[0086] d(A i , B j )=(A i -B j ) 2

[0087] Among them, A i is the coordinate of the i-th scattered point in time series A, B j is the coordinate of the jth scattered point in time series B.

[0088] Each matrix element in the matrix grid represents a scattered point A i With scatter point B j alignment.

[0089] In order to determine the target scattered points after the alignment process, the DP algorithm can be used for reduction. Specifically, in the obtained matrix grid, there are multiple paths from the (1, 1)th matrix element to the (n, m)th matrix element. The sum of the matrix elements passed by each path is calculated, and the path corresponding to the minimum sum is determined as the shortest path. Among them, the scattered points of the two learning sequences corresponding to each matrix element contained in the shortest path are the target scattered points after alignment. This shortest path is defined as the warping path and is represented by w. The pth element of w is defined as w(p) = (i, j) p , w(p) is used to represent the aligned target scatter points when the target student A and the target student B learn the p-th chapter video, that is, it defines the mapping between time series A and time series B.

[0090] S4: Calculate the relative distance between any two target students when they study the target online course based on the aligned target scatter points when any two target students study the same chapter video.

[0091] Furthermore, in step S4, the relative distance between any two target students when learning the target online course can be calculated using the following formula:

[0092]

[0093] Wherein, D(A, B) is the relative distance between target student A and target student B when they study the target online course, taking the minimum value in the function F[·]; P is the number of chapter videos contained in the target online course; w(p) is the aligned target scatter point when target student A and target student B study the pth chapter video; a(p) and b(p) are the viewing time of target student A and target student B when they study the pth chapter video respectively; d(a(p), b(p)) is the difference in viewing time when target student A and target student B study the pth chapter video.

[0094] In order to solve the above formula, we can set the initial condition g(1,1)=2d(1,1), where g(i,j)=min[g(i,j-1)+d(i,j),g(i-1,j-1)+2d(i,j),g(i-1,j)+d(i,j)], where the window restriction condition is: jr≤i≤j+r, r is the intermediate parameter. Then the relative distance D(A,B)=g(i,j) / (i+j). Please refer to the flowchart of the solution process. Figure 2 .in Figure 2 Here, I is the maximum value of i, and J is the maximum value of j.

[0095] Next, for step 106, based on the relative distance between any two target students when learning the target online course, the learning behaviors of learning the target online course are clustered to obtain multiple types of learning behaviors corresponding to learning the target online course.

[0096] After obtaining the relative distance between any two target students learning the target online course, the K-nearest neighbor algorithm (KNN) can be used for clustering to obtain several types of exemplary learning behaviors of excellent students. In this embodiment, an automatic iteration method can be used to allow the machine to automatically select a suitable K value according to the error value. Specifically, step 106 may include the following steps R1 to R3:

[0097] R1: Determine the adaptive K value.

[0098] R2: Take each target student as the center, sort the relative distances between other target students and the target student at the center according to size, and select the other target students corresponding to the first K smallest relative distances and the target student at the center as one category;

[0099] R3: Determine the classification results of all target students according to the categories with the highest number of occurrences, and obtain the learning behaviors of the target students included in each category in learning the target online course.

[0100] When determining the K value, the K value can be initialized to 1, and then the above steps are used to perform automatic iterative clustering, and the K value with the smallest error value is selected as the adaptive K value. After clustering using the adaptive K value, the final classification result is obtained.

[0101] Assume that five types of learning behaviors can be obtained. The first type of learning behavior has more repeated viewings at the 3rd to 5th minute viewing time point of the first chapter video, and the second type of learning behavior has more repeated viewings at the 10th to 15th minute viewing time point of the first chapter video...

[0102] Furthermore, since the same type of learning behavior includes the learning behaviors of multiple target students, due to the high similarity, there are still differences. Therefore, the learning behavior of one of the target students can be selected as the learning behavior of this type, or the learning behavior of this type can be determined by calculating the mean. More preferably, the learning behavior corresponding to the target student at the cluster center point in this type can be selected as the learning behavior of this type.

[0103] Finally, for step 108, a learning behavior suitable for the target online course is determined based on the obtained multiple types of learning behaviors.

[0104] Since multiple types of learning behaviors are all based on the learning behaviors of excellent students, when determining the learning behaviors suitable for the target online course, any type of learning behavior can be determined as the learning behaviors suitable for the target online course. However, considering that different types of learning behaviors have different focuses on the target online course, in order to provide students with more comprehensive guidance on learning behaviors, in one embodiment, the learning behaviors suitable for the target online course can be determined in the following manner: according to each type of learning behavior, the number of views corresponding to each viewing time point of the target online course is determined, and the number of views corresponding to each viewing time point of the target online course is determined as the learning behaviors suitable for the target online course.

[0105] Continuing with the above-mentioned first and second types of learning behaviors as examples, since the first type of learning behavior has a large number of repeated viewings at the 3rd to 5th minute viewing time point of the first chapter video, and the second type of learning behavior has a large number of repeated viewings at the 10th to 15th minute viewing time point of the first chapter video, then the two learning behaviors can be taken as a union, that is, the 3rd to 5th minute viewing time point and the 10th to 15th minute viewing time point of the first chapter video are both determined as key points of the course, to remind students that these two parts of the course clips need to be focused on and can be watched repeatedly. In this way, it can ensure that the determined learning behaviors are more comprehensive, and can give more prompts to students who subsequently study the course, and assist students in learning the target online course.

[0106] In one embodiment of the present invention, after determining the learning behavior suitable for the target online course, the method may further include: providing prompts to students who are learning the target online course according to the learning behavior suitable for the target online course. For example, real-time prompts during the learning process, reminders before learning, etc. Feedback may also be provided to educators so that they can understand the problems and difficulties in students' learning, which can be used as a basis for subsequent course improvements.

[0107] Based on the data provided by a MOOC platform, this scheme is used for experimental analysis and verification. The data of three real online courses are used as the basis for testing. Multiple experiments are conducted to evaluate the error rate, operation time, availability and other parameters of the scheme. The experiments are conducted on a Windows machine with a 64-bit 4-core 2.1GHz AMD(R) Ryzen 5 3500u CPU and 8GB memory, and the algorithm is implemented in Python.

[0108] Error rate: When determining the learning behavior of three online courses, the error rates are all lower than 10%, namely 9.6%, 4.7%, and 8.2% respectively. It can be seen that the error rate of the final KNN algorithm is quite low.

[0109] Calculation time: The number of data items corresponding to the three online courses are 324351, 195398, and 85462, respectively. The final calculation time is 9.82s, 6.73s, and 2.54s, respectively. It can be seen that it can meet the basic requirements of providing real-time academic guidance.

[0110] Usability: After testing, the completion rate and average score of students who used the learning behaviors determined by the program as a learning aid were much higher than those of students who did not use it, proving that the program has the effect of increasing learning enthusiasm and improving students' academic performance.

[0111] like Figure 3 , Figure 4 As shown, an embodiment of the present invention provides an online course learning behavior determination device. The device embodiment can be implemented by software, or by hardware or a combination of software and hardware. From the hardware level, Figure 3 As shown, it is a hardware architecture diagram of a computing device where an online course learning behavior determination device provided by an embodiment of the present invention is located, except Figure 3 In addition to the processor, memory, network interface, and non-volatile memory shown in the figure, the computing device in which the device is located in the embodiment may also generally include other hardware, such as a forwarding chip responsible for processing messages, etc. Taking software implementation as an example, Figure 4 As shown, as a device in a logical sense, the CPU of the computing device in which it is located reads the corresponding computer program in the non-volatile memory into the memory and runs it. This embodiment provides an online course learning behavior determination device, including:

[0112] The historical data acquisition unit 401 is used to acquire historical data related to the target online course; the historical data at least includes information about students who study the target online course and behavior information about the students in the process of studying the target online course;

[0113] A target student determination unit 402 is used to determine, among the students studying the target online course, target students whose learning behaviors for the target online course can be referenced according to the historical data;

[0114] A relative distance determination unit 403 is used to determine the relative distance between any two target students when they are learning the target online course according to the behavior information of each target student in the process of learning the target online course;

[0115] A clustering unit 404 is used to cluster the learning behaviors of learning the target online course according to the relative distance between any two target students when learning the target online course, so as to obtain a plurality of types of learning behaviors corresponding to learning the target online course;

[0116] The learning behavior determining unit 405 is used to determine a learning behavior suitable for the target online course according to the obtained multiple types of learning behaviors.

[0117] In one embodiment of the present invention, the target student determination unit 402 is specifically configured to determine the student as the target student if it is determined that the student's test score for the target online course meets a preset level and / or it is determined that the student's learning activity for the target online course meets a preset activity.

[0118] In one embodiment of the present invention, the relative distance determination unit 403 is specifically used to determine each chapter video included in the target online course; abstract the behavior information of each target student learning each chapter video into discrete scatter points, and generate a corresponding time series based on the abstracted scatter points; according to the time series corresponding to each target student learning each chapter video, use a dynamic time warping algorithm to align the time series corresponding to any two target students learning the same chapter video to obtain the aligned target scatter points when any two target students learn the same chapter video; according to the aligned target scatter points when any two target students learn the same chapter video, calculate the relative distance between any two target students learning the target online course.

[0119] In one embodiment of the present invention, when the relative distance determination unit 403 performs the step of abstracting the behavior information of each target student in learning each chapter video into discrete scatter points and generating a corresponding time series based on the abstracted scatter points, it specifically includes: determining the number of views of each target student on the chapter video at each viewing time point based on the behavior information of each target student on the chapter video; the viewing time point is not less than 0 and not greater than the maximum playback time point of the chapter video; the viewing time point and the corresponding number of views are abstracted into a scatter point, and then, for each target student, a time series for learning the chapter video is obtained.

[0120] In one embodiment of the present invention, the relative distance determination unit 403 performs the alignment processing of the time series corresponding to any two target students learning the same chapter video using the dynamic time warping algorithm to obtain the aligned target scatter points when any two target students learn the same chapter video, specifically including: constructing an n×m matrix grid for the time series corresponding to any two target students learning the same chapter video; wherein n and m are the number of scatter points contained in the time series corresponding to the two target students learning the same chapter video; the (i, j)th matrix element in the matrix grid is the distance between the i-th scatter point in the time series corresponding to one of the target students and the j-th scatter point in the time series corresponding to the other target student; for the matrix grid, determining the shortest path through the matrix elements from the (1, 1)th matrix element to the (n, m)th matrix element; and determining the scatter points of the two learning sequences corresponding to each matrix element contained in the shortest path as the aligned target scatter points.

[0121] In one embodiment of the present invention, when the relative distance determining unit 403 performs the calculation to obtain the relative distance between any two target students when learning the target online course, the method specifically includes: using the following formula to calculate the relative distance between any two target students when learning the target online course:

[0122]

[0123] Wherein, D(A, B) is the relative distance between target student A and target student B when they study the target online course, taking the minimum value in the function F[·]; P is the number of chapter videos contained in the target online course; w(p) is the aligned target scatter point when target student A and target student B study the pth chapter video; a(p) and b(p) are the viewing time of target student A and target student B when they study the pth chapter video respectively; d(a(p), b(p)) is the difference in viewing time when target student A and target student B study the pth chapter video.

[0124] In one embodiment of the present invention, the clustering unit is specifically used to determine an adaptive K value; taking each target student as the center, the relative distances between other target students and the target student at the center are sorted according to size, and the other target students corresponding to the first K smallest relative distances and the target student at the center are selected as one category; according to the category with the largest number of occurrences, the result of classifying all target students is determined, and the learning behaviors of the target students included in each category in learning the target online course are obtained.

[0125] In one embodiment of the present invention, the learning behavior determination unit is specifically used to: determine the number of views corresponding to each viewing time point of the target online course according to each type of learning behavior, and determine the number of views corresponding to each viewing time point of the target online course as a learning behavior suitable for the target online course;

[0126] In one embodiment of the present invention, the learning behavior determination unit is further configured to prompt students studying the target online course according to the learning behavior suitable for the target online course.

[0127] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on an online course learning behavior determination device. In other embodiments of the present invention, an online course learning behavior determination device may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0128] The information interaction, execution process and other contents between the modules in the above-mentioned device are based on the same concept as the embodiment of the method of the present invention. For the specific contents, please refer to the description in the embodiment of the method of the present invention, and no further description is given here.

[0129] An embodiment of the present invention further provides a computing device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, a method for determining online course learning behavior in any embodiment of the present invention is implemented.

[0130] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the processor executes a method for determining online course learning behavior in any embodiment of the present invention.

[0131] Specifically, a system or device equipped with a storage medium can be provided, on which software program code that implements the functions of any of the above-mentioned embodiments is stored, and a computer (or CPU or MPU) of the system or device can be enabled to read and execute the program code stored in the storage medium.

[0132] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0133] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer by a communication network.

[0134] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0135] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion module connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or expansion module is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0136] It should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical factors in the process, method, article or device including the elements.

[0137] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining learning behavior in online courses, It is characterized in that include: Obtain historical data related to the target online course; The historical data at least includes information about students who study the target online course and information about students’ behaviors during the process of studying the target online course; determining, among students studying the target online course, target students whose learning behaviors for the target online course can be used as references according to the historical data; Determine the relative distance between any two target students when they are learning the target online course according to the behavior information of each target student in the process of learning the target online course; Clustering the learning behaviors of learning the target online course according to the relative distance between any two target students when learning the target online course, to obtain multiple types of learning behaviors corresponding to learning the target online course; Determining a learning behavior suitable for the target online course according to the obtained multiple types of learning behaviors; The behavior information includes the chapter videos, viewing time points, the number of views, viewing duration and learning time of each viewing time point of the target online course studied by the student; wherein the viewing time point refers to the time point in the chapter video, and the learning time refers to natural time; The determining of target students whose learning behaviors for the target online course can be referenced includes: If it is determined that the student's test score for the target online course meets a preset level, and / or it is determined that the student's learning activity for the target online course meets a preset activity, the student is determined as the target student; The determining of the relative distance between any two target students when studying the target online course comprises: Determine the individual chapter videos included in the target online course; Abstracting the behavior information of each target student learning each chapter video into discrete scattered points, and generating corresponding time series according to the abstracted scattered points; According to the time sequence corresponding to each target student learning each chapter video, the time sequences corresponding to any two target students learning the same chapter video are aligned using a dynamic time warping algorithm to obtain aligned target scatter points when any two target students learn the same chapter video; According to the aligned target scatter points when any two target students learn the same chapter video, the relative distance between any two target students when learning the target online course is calculated; The step of abstracting the behavior information of each target student learning each chapter video into discrete scattered points, and generating a corresponding time series according to the abstracted scattered points, includes: According to the behavior information of each target student on the chapter video, the number of views of each target student on the chapter video at each viewing time point is determined; the viewing time point is not less than 0 and not greater than the maximum playback time point of the chapter video; The viewing time point and the corresponding viewing number are abstracted into a scattered point, and then for each of the target students, the time series of learning the video of the chapter is obtained; The method uses a dynamic time warping algorithm to align the time series corresponding to any two target students learning the same chapter video to obtain the aligned target scattered points when any two target students learn the same chapter video, including: Construct an n×m matrix grid for the time series corresponding to any two target students when they learn the same chapter video; wherein n and m are the number of scattered points contained in the time series corresponding to the two target students when they learn the same chapter video; the (i, j)th matrix element in the matrix grid is the distance between the i-th scattered point in the time series corresponding to one of the target students and the j-th scattered point in the time series corresponding to the other target student; For the matrix grid, determining the shortest path from the (1,1)th matrix element to the (n,m)th matrix element through the matrix elements; Determine the scattered points of the two learning sequences corresponding to each matrix element included in the shortest path as the aligned target scattered points; The calculation to obtain the relative distance between any two target students when studying the target online course includes: The relative distance between any two target students when learning the target online course is calculated using the following formula: Wherein, D(A,B) is the relative distance between target student A and target student B when learning the target online course, taking the minimum value in the function F[·]; P is the number of chapter videos contained in the target online course; w(p) is the aligned target scatter point when target student A and target student B learn the pth chapter video; a(p) and b(p) are the viewing time of target student A and target student B learning the pth chapter video respectively; d(a(p),b(p)) is the difference in viewing time when target student A and target student B learn the pth chapter video.

2. The method according to claim 1, It is characterized in that The learning behaviors of learning the target online course are clustered according to the relative distance between any two target students when learning the target online course, to obtain multiple types of learning behaviors corresponding to learning the target online course, including: Determine the adaptive K value; Take each target student as the center, sort the relative distances between other target students and the target student at the center by size, and select the other target students corresponding to the first K smallest relative distances and the target student at the center as one category; According to the categories with the highest number of occurrences, the results of classifying all target students are determined, and the learning behaviors of the target students included in each category in learning the target online course are obtained; and / or, Determining a learning behavior suitable for the target online course based on the obtained multiple types of learning behaviors includes: According to each type of learning behavior, determining the number of views corresponding to each viewing time point of the target online course, and determining the number of views corresponding to each viewing time point of the target online course as a learning behavior suitable for the target online course; and / or, After determining the learning behavior suitable for the target online course, the method further includes: providing prompts to students studying the target online course according to the learning behavior suitable for the target online course.

3. A device for determining online course learning behavior, It is characterized in that include: A historical data acquisition unit, used for acquiring historical data related to a target online course; The historical data at least includes information about students who study the target online course and information about students’ behaviors during the process of studying the target online course; a target student determination unit, configured to determine, based on the historical data, a target student whose learning behavior for the target online course can be used as a reference among students studying the target online course; A relative distance determination unit, configured to determine the relative distance between any two target students when they are learning the target online course according to the behavior information of each target student in the process of learning the target online course; A clustering unit, configured to cluster the learning behaviors of learning the target online course according to the relative distance between any two target students when learning the target online course, and obtain a plurality of types of learning behaviors corresponding to learning the target online course; A learning behavior determination unit, configured to determine a learning behavior suitable for the target online course according to the obtained multiple types of learning behaviors; The behavior information includes the chapter videos, viewing time points, the number of views, viewing duration and learning time of each viewing time point of the target online course studied by the student; wherein the viewing time point refers to the time point in the chapter video, and the learning time refers to natural time; The target student determination unit is specifically configured to determine the student as the target student if it is determined that the student's test score for the target online course meets a preset level, and / or it is determined that the student's learning activity for the target online course meets a preset activity; The relative distance determination unit is specifically used to determine each chapter video included in the target online course; abstract the behavior information of each target student learning each chapter video into discrete scattered points, and generate a corresponding time series based on the abstracted scattered points; according to the time series corresponding to each target student learning each chapter video, use a dynamic time warping algorithm to align the time series corresponding to any two target students learning the same chapter video, and obtain the aligned target scattered points when any two target students learn the same chapter video; according to the aligned target scattered points when any two target students learn the same chapter video, calculate the relative distance when any two target students learn the target online course; When the relative distance determination unit performs the step of abstracting the behavior information of each target student learning each chapter video into discrete scattered points and generating a corresponding time series based on the abstracted scattered points, it specifically includes: determining the number of views of each target student on the chapter video at each viewing time point according to the behavior information of each target student on the chapter video; the viewing time point is not less than 0 and not greater than the maximum playback time point of the chapter video; the viewing time point and the corresponding number of views are abstracted into a scattered point, and then, for each target student, a time series when learning the chapter video is obtained; When the relative distance determination unit performs the alignment processing of the time series corresponding to any two target students learning the same chapter video by using the dynamic time warping algorithm to obtain the aligned target scattered points when any two target students learn the same chapter video, it specifically includes: constructing an n×m matrix grid for the time series corresponding to any two target students learning the same chapter video; wherein n and m are the number of scattered points contained in the time series corresponding to the two target students learning the same chapter video; the (i, j)th matrix element in the matrix grid is the distance between the i-th scattered point in the time series corresponding to one of the target students and the j-th scattered point in the time series corresponding to the other target student; for the matrix grid, determining the shortest path through the matrix elements from the (1, 1)th matrix element to the (n, m)th matrix element; determining the scattered points of the two learning sequences corresponding to each matrix element contained in the shortest path as the aligned target scattered points; When the relative distance determination unit performs the calculation to obtain the relative distance between any two target students when learning the target online course, the relative distance determination unit specifically includes: using the following formula to calculate the relative distance between any two target students when learning the target online course: Wherein, D(A,B) is the relative distance between target student A and target student B when learning the target online course, taking the minimum value in the function F[·]; P is the number of chapter videos contained in the target online course; w(p) is the aligned target scatter point when target student A and target student B learn the pth chapter video; a(p) and b(p) are the viewing time of target student A and target student B learning the pth chapter video respectively; d(a(p),b(p)) is the difference in viewing time when target student A and target student B learn the pth chapter video.

4. A computing device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method according to any one of claims 1 to 2 is implemented.

5. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the method according to any one of claims 1 to 2.