CN107133894A - On-line study group technology based on Complex Networks Theory - Google Patents
On-line study group technology based on Complex Networks Theory Download PDFInfo
- Publication number
- CN107133894A CN107133894A CN201710321172.2A CN201710321172A CN107133894A CN 107133894 A CN107133894 A CN 107133894A CN 201710321172 A CN201710321172 A CN 201710321172A CN 107133894 A CN107133894 A CN 107133894A
- Authority
- CN
- China
- Prior art keywords
- learner
- network
- learning
- euclidean distance
- composite
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Electrically Operated Instructional Devices (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of on-line study group technology based on Complex Networks Theory, comprise the following steps：In extensive on-line study system, the data acquisition system for being capable of various dimensions reflection learner's personal traits is obtained；Learner's school grade is obtained, and combines the data acquisition system of the reflection learner's personal traits obtained and builds compound Euclidean distance；Each learner is considered as to a node in learner's network, learner's network is built；Each node for having side to connect is merged successively using greedy algorithm, automatic cluster is divided into study group.The present invention realizes the scientific and reasonable quick division of on-line study group, simplifies the workload of on-line study group division, improves efficiency, validity and applicability.
Description
Technical Field
The invention relates to the field of online education, in particular to an online learning grouping method based on a complex network theory.
Background
Online Open Courses such as MOOC (Massive Open Online Courses) are developed vigorously and also have some defects, large-scale Online Courses and the characteristics of space-time separation of teachers and students make targeted guidance lack, so that the participation and interaction of learners are insufficient, learning has an autism, learning experience lacks integrity, and the like, and the effect of Online learning is finally influenced.
Cooperative learning is a major trend of online education, and collaborative innovative teaching for the purpose of promoting cooperative learning among learners and conversation with teachers to achieve visual field fusion is an important characteristic of online education. Under the online learning situation of the teacher and the student with space-time separation, the cooperative learning emphasizes the interaction, communication and understanding among all the main bodies, so that the learning loneliness sense is eliminated and the growth sense of self value is obtained, and the division of the learning groups is the primary link of the cooperative learning.
The current online learning grouping method mainly depends on manual assignment or random grouping of teachers, but the manual assignment has large workload and difficult management, and the random grouping lacks effective analysis on characteristics of learners, and has large randomness and poor applicability.
In view of the above, in a large-scale online learning system, it is urgently needed to simplify the workload of online learning group division so as to improve the efficiency, effectiveness and applicability thereof.
Disclosure of Invention
The invention aims to solve the technical problem of simplifying the workload of online learning group division in a large-scale online learning system so as to improve the efficiency, effectiveness and applicability of the online learning system.
In order to solve the technical problems, the technical scheme adopted by the invention is to provide an online learning grouping method based on a complex network theory, which comprises the following steps:
in a large-scale online learning system, acquiring a data set capable of reflecting the personality traits of a learner in multiple dimensions;
acquiring the learning score of the learner, and constructing a composite Euclidean distance by combining the acquired data set reflecting the personality traits of the learner;
each learner is regarded as a node in the learner network, and the learner network is constructed;
and combining the nodes connected with the edges in sequence by using a greedy algorithm, and automatically clustering and dividing into learning groups.
In the above-described technical solution, the node having the largest number of nodes in the learning group is used as the learning leader of the learning group.
In the above technical solution, each learner is regarded as a node in a learner network, and the construction of the learner network specifically includes:
setting a composite Euclidean distance threshold value between nodes, and when the composite Euclidean distance between two nodes is smaller than or equal to the composite Euclidean distance threshold value, considering that the two nodes are connected with edges;
and when the composite Euclidean distance between the two nodes is greater than the composite Euclidean distance threshold value, the two nodes are considered to have no connection relation.
In the technical scheme, the composite Euclidean distance threshold is N times of the average value of the composite Euclidean distances of all nodes in the learner network, and N is more than or equal to 0.3 and less than or equal to 1.5.
In the above technical solution, the composite euclidean distance is expressed as:
wherein D is a composite Euclidean distance; d1 is the Euclidean distance of the learner's learning achievement; d2 is the Euclidean distance of the learner personality; α is the weight of d 1; β is the weight of d 2; α + β ═ 1.
In the technical scheme, the nodes connected with edges are combined in sequence by using a greedy algorithm, and the automatic clustering is divided into learning groups, and the method specifically comprises the following steps:
each node in the learner network is regarded as a community;
merging the communities connected with the edges in sequence;
calculating the modularity of the new community formed after combination until the modularity of the whole learner network is maximum, wherein the obtained new community is a learning group;
according to the principle of a greedy algorithm, the merging direction is carried out along the direction of maximum modularity increase or minimum modularity decrease of the learner network every time, a certain number of communities are reduced in each merging, and meanwhile, the structure of the learner network is correspondingly updated until the nodes which are connected with one another are merged into one community.
In the above technical solution, the greedy algorithm specifically includes the following steps:
defining an array symmetric matrix E ═ E of dimension n × nij]；
Wherein n is the number of communities divided by the learner network, and initially, n is the number of nodes in the learner network;
defining the sum of elements in each row or each column of the array symmetric matrix E as ap＝∑jeij；
Wherein,m is the total number of edges of the learner's network, kpIs the modularity of node p;
and combining the communities connected with the edges in sequence, calculating the modularity of the new community, and repeating the processes until the modularity of the whole learner network is maximum.
In the above technical solution, the activity of the learner is represented by a point degree, the centrality of the point degree of the learner is a combination of a point out degree and a point in degree, and the activity comprehensive evaluation value of the ith learner is as follows:
Ai＝αi×outDi+βi×inDi；
wherein Ai represents the comprehensive evaluation value of the activity of the ith learner; outDiAnd inDiRespectively representing the point-out degree and the point-in degree of the ith learner αiAnd βiWeights representing the point-out and point-in of the ith learner, respectively, αi+βi＝1。
In the above technical solution, the influence of the learner is expressed by the recenterness and the pitch-centralness, and the comprehensive evaluation value of the influence of the ith learner is as follows:
Ii＝γi×Cc+μi×Cb；
wherein, IiRepresenting the comprehensive evaluation value of the influence of the ith learner; ccAnd CbRespectively representing the approach centrality and the distance centrality of the ith learner; gamma rayiAnd muiWeights, γ, representing the recenterness and distance-centrality of the ith learner, respectivelyi+μi＝1。
In the teaching and management of large-scale online courses, the invention adopts a multi-dimensional data clustering algorithm, finds learners with similar learning styles and characteristics from massive learner big data with isomerism, multi-dimension and mass without manual intervention, and automatically clusters the learners into an online learning group, thereby realizing scientific and reasonable rapid division of the online learning group, simplifying the workload of division of the online learning group, and improving the efficiency, the effectiveness and the applicability.
Drawings
FIG. 1 is a flow chart of an online learning grouping method based on a complex network theory according to the present invention;
FIG. 2 is a diagram of a process for constructing a data set of personality traits of a learner according to the present invention;
FIG. 3 is a diagram illustrating the community division in the learner network according to the present invention.
Detailed Description
In the teaching and management of large-scale online courses, in order to simplify the workload of online learning group division and improve the efficiency, effectiveness and applicability of online learning group division, the embodiment of the invention provides an online learning grouping method based on a complex network theory.
Meanwhile, by combining a complex network theory and community characteristics thereof, by utilizing the concept of betweenness (reflecting the action and influence of corresponding nodes or edges in the whole network), using the node betweenness as an index for finding a learning leader, and selecting the node with the maximum node betweenness in an online learning group (each learner is regarded as one node in the learning network) as the learning leader, wherein the learning leader is an individual with high liveness, excellent performance and guidance on learning interaction and deep learning in the online learning process and can play a role in teaching assistance, thereby realizing the teaching effect of the grand brothers.
The invention is described in detail below with reference to the drawings and the detailed description.
As shown in fig. 1, an online learning grouping method based on a complex network theory provided in an embodiment of the present invention includes the following steps:
s10, in the large-scale online learning system, acquiring learning behavior Data-C capable of reflecting the cognitive style of the learner, and combining preset learning style preference Data-D to acquire a Data set capable of reflecting the personality traits of the learner in multiple dimensions.
As shown in FIG. 2, the time and frequency of the learner accessing the forum, the posting amount/reading amount of the forum, and the time and frequency of the learner browsing the concrete/abstract/example/knowledge tree/text/video/diagram (image) in the online learning system are collected and compared with the whole learning duration to determine the behavior pattern of the learner, and the behavior pattern is determined through a learning style scale Ai＝αi×outDi+βi×inDiAnd carrying out quantitative calculation, and classifying the obtained learner cognitive style according to an active type/meditation type, an apprehension type/intuition type, a visual type/speech type, a sequence type/comprehensive type and the like. The personality traits of the learners include extroversion, hommization, responsibility, nervousness, experience openness and the like.
And S11, acquiring the learning achievement of the learner, and constructing the composite Euclidean distance by combining the acquired data set reflecting the personality traits of the learner.
The method for acquiring the learning achievement of the learner specifically comprises the following steps: the method comprises the steps of obtaining the course evaluation (unit test, learning duration, participation frequency and quality, mutual evaluation frequency and quality, liveness, posting and replying amount and the like) and the final evaluation of learners, and weighting to form a total score which is only used for reflecting the learning effect of online learners.
The composite Euclidean distance comprises the Euclidean distance of the learning achievement of the learner and the Euclidean distance of the personality traits of the learner, and is expressed as follows:
wherein D is a composite Euclidean distance; d1 is the Euclidean distance of the learner's learning achievement; d2 is the Euclidean distance of the personality trait of the learner; α is the weight of Euclidean distance d 1; β is the weight of the euclidean distance d 2; α + β ═ 1.
The above dimension of euclidean distance d1 of the learner's learning achievement includes: Data-A-second classroom information such as course selection records, participating communities, competitions, lectures and book borrowing records; and Data-B-on-line learning response accuracy, time for solving problems, modifying times, total learning duration, note making times, posting number, courseware browsing achievement, in-class discussion achievement, objective exercise achievement, subjective exercise achievement, on-line achievement and off-line achievement.
The above dimension of the euclidean distance d2 of the learner personality trait includes: Data-C from learning behavior-frequency and time and scale of browsing the content of concrete/abstract/instance/knowledge tree/text/video/chart; the number of friends (for the one-way linked microblogs, the number is divided into the number of concerned and the number of concerned), short messages, comments, photos/albums, activities, hobbies, music, movies, cognitive styles and other contents; the number of parameters such as work related words, exclamation marks and the like in the text information; and from preset learning style preference Data-D.
In order to eliminate dimension influence among all indexes and solve comparability among the indexes, a Min-max Normalization method is adopted to normalize the Data-A, the Data-B, the learning behavior Data-C and the preset learning style preference Data-D, so that the value of the composite Euclidean distance D is mapped between [0-1 ].
And S12, constructing a learner network.
Specifically, each learner is regarded as a node in the learner network, and a composite Euclidean distance threshold value between the nodes is set, wherein N times of the average value of the composite Euclidean distances of all the nodes is selected as the composite Euclidean distance threshold value, N is more than or equal to 0.3 and less than or equal to 1.5, and when the composite Euclidean distance between the nodes is less than or equal to the composite Euclidean distance threshold value, the nodes are regarded as being connected with edges; and when the composite Euclidean distance between the nodes is larger than the composite Euclidean distance threshold value, the nodes are considered to have no connection.
And S13, sequentially merging the nodes connected with the edges by using a greedy algorithm (Fast community algorithm), and automatically clustering and dividing into learning groups.
As shown in fig. 3, which is a schematic diagram illustrating the division of communities in the learner network, first, each node in the learner network is regarded as a community; secondly, merging the communities connected with the edges in sequence, wherein the merging direction is performed along the direction of the maximum modularity increase or the minimum modularity decrease of the learner network according to the greedy algorithm principle, a certain number of communities are reduced in each merging, and meanwhile, the structure of the learner network is correspondingly updated until the nodes connected with the edges are merged into one community; and finally, recalculating the modularity of the new community formed after merging until the modularity of the whole learner network is maximum. In the present embodiment, a community can be considered as a learning group.
The specific process of Fast community algorithm is as follows:
defining an array symmetric matrix E ═ E of dimension n × nij]；
Wherein n is the number of communities divided by the learner network, and initially, n is the number of nodes in the learner network;
defining the sum of elements in each row or each column of the array symmetric matrix E as ap＝∑jeij；
Wherein,m is the total number of edges of the learner's network, kpIs the modularity of node p;
and combining the communities connected with the edges in sequence, calculating the modularity Q value of the new community, and repeating the process until the modularity Q value of the whole learner network is maximum.
In the learning groups, the node with the largest number of node intermediaries in each learning group is used as the learning leader of the learning group, and the learning leader is an individual with high liveness and excellent performance and has a leading function on learning interaction and deep learning in the online learning process.
The liveness is expressed by the degree of a point (degree center), the centricity of the degree of a learner is the combination of the degree of a point and the degree of a point, and the comprehensive evaluation value Ai of the liveness of the ith learner is defined as:
Ai＝αi×outDi+βi×inDi；
wherein outDiAnd inDiRespectively representing the point-out degree and the point-in degree of the ith learner αiAnd βiWeights representing the point-out and point-in of the ith learner, respectively, αi+βi＝1。
The influence is represented by the degree of approaching the center and the degree of spacing center, and the influence of the ith learner is comprehensively evaluated by the evaluation value IiIs defined as:
Ii＝γi×Cc+μi×Cb；
wherein, CcAnd CbRespectively representing the approach centrality and the distance centrality of the ith learner; gamma rayiAnd muiWeights, γ, representing the recenterness and distance-centrality of the ith learner, respectivelyi+μi＝1。
The present invention is not limited to the above-mentioned preferred embodiments, and any structural changes made by anyone in the light of the present invention, all the technical solutions similar or similar to the present invention, fall within the protection scope of the present invention.
Claims (9)
1. The online learning grouping method based on the complex network theory is characterized by comprising the following steps of:
in a large-scale online learning system, acquiring a data set capable of reflecting the personality traits of a learner in multiple dimensions;
acquiring the learning score of the learner, and constructing a composite Euclidean distance by combining the acquired data set reflecting the personality traits of the learner;
each learner is regarded as a node in the learner network, and the learner network is constructed;
and combining the nodes connected with the edges in sequence by using a greedy algorithm, and automatically clustering and dividing into learning groups.
2. The method of claim 1, wherein a node with the largest node betweenness in a learning group is used as a learning leader of the learning group.
3. The method of claim 1, wherein each learner is considered as a node in a learner network, and the learner network is constructed by:
setting a composite Euclidean distance threshold value between nodes, and when the composite Euclidean distance between two nodes is smaller than or equal to the composite Euclidean distance threshold value, considering that the two nodes are connected with edges;
and when the composite Euclidean distance between the two nodes is greater than the composite Euclidean distance threshold value, the two nodes are considered to have no connection relation.
4. The method of claim 3 wherein the composite Euclidean distance threshold is N times the average of the composite Euclidean distances of all nodes in the learner's network, 0.3 ≦ N ≦ 1.5.
5. The method of claim 1, wherein the composite euclidean distance is expressed as:
<mrow> <mi>D</mi> <mo>=</mo> <mi>&alpha;</mi> <mfrac> <mn>1</mn> <msub> <mi>d</mi> <mn>1</mn> </msub> </mfrac> <mo>+</mo> <msub> <mi>&beta;d</mi> <mn>2</mn> </msub> <mo>;</mo> </mrow>
wherein D is a composite Euclidean distance; d1 is the Euclidean distance of the learner's learning achievement; d2 is the Euclidean distance of the learner personality; α is the weight of d 1; β is the weight of d 2; α + β ═ 1.
6. The method of claim 1, wherein the nodes connected by edges are merged in turn by a greedy algorithm, and the automatic clustering is divided into learning groups, and the method specifically comprises the following steps:
each node in the learner network is regarded as a community;
merging the communities connected with the edges in sequence;
calculating the modularity of the new community formed after combination until the modularity of the whole learner network is maximum, wherein the obtained new community is a learning group;
according to the principle of a greedy algorithm, the merging direction is carried out along the direction of maximum modularity increase or minimum modularity decrease of the learner network every time, a certain number of communities are reduced in each merging, and meanwhile, the structure of the learner network is correspondingly updated until the nodes which are connected with one another are merged into one community.
7. The method of claim 6, wherein the greedy algorithm is implemented as follows:
defining an array symmetric matrix E ═ E of dimension n × nij]；
Wherein n is the number of communities divided by the learner network, and initially, n is the number of nodes in the learner network;
defining the sum of elements in each row or each column of the array symmetric matrix E as ap＝∑jeij；
Wherein,m is the total number of edges of the learner's network, kpIs the modularity of node p;
and combining the communities connected with the edges in sequence, calculating the modularity of the new community, and repeating the processes until the modularity of the whole learner network is maximum.
8. The method of claim 2, wherein the activeness of the learner is represented by a point degree, the centricity of the point degree of the learner is a combination of a point-out degree and a point-in degree, and the activity comprehensive evaluation value of the ith learner is as follows:
Ai＝αi×outDi+βi×inDi；
wherein Ai represents the comprehensive evaluation value of the activity of the ith learner; outDiAnd inDiRespectively representing the point-out degree and the point-in degree of the ith learner αiAnd βiWeights representing the point-out and point-in of the ith learner, respectively, αi+βi＝1。
9. The method of claim 2, wherein the learner's influence is represented by a recenterness and a distance-centralness, and the integrated evaluation value of the influence of the ith learner is as follows:
Ii＝γi×Cc+μi×Cb；
wherein, IiRepresenting the comprehensive evaluation value of the influence of the ith learner; ccAnd CbRespectively representing the approach centrality and the distance centrality of the ith learner; gamma rayiAnd muiWeights, γ, representing the recenterness and distance-centrality of the ith learner, respectivelyi+μi＝1。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710321172.2A CN107133894B (en) | 2017-05-09 | 2017-05-09 | Online learning grouping method based on complex network theory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710321172.2A CN107133894B (en) | 2017-05-09 | 2017-05-09 | Online learning grouping method based on complex network theory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107133894A true CN107133894A (en) | 2017-09-05 |
| CN107133894B CN107133894B (en) | 2021-05-25 |
Family
ID=59732686
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710321172.2A Active CN107133894B (en) | 2017-05-09 | 2017-05-09 | Online learning grouping method based on complex network theory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107133894B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109712011A (en) * | 2017-10-25 | 2019-05-03 | 北京京东尚科信息技术有限公司 | Community discovery method and device |
| CN110033239A (en) * | 2019-04-16 | 2019-07-19 | 暨南大学 | Row's test method, device, equipment and medium based on greedy algorithm |
| CN110738197A (en) * | 2019-11-12 | 2020-01-31 | 上海乂学教育科技有限公司 | Learning state evaluation model construction method |
| CN111063242A (en) * | 2020-01-14 | 2020-04-24 | 重庆工业职业技术学院 | Accounting learning method and accounting learning device |
| CN112115357A (en) * | 2020-09-11 | 2020-12-22 | 华中师范大学 | Online course forum interaction mode identification method and system |
| CN114580912A (en) * | 2022-03-07 | 2022-06-03 | 浙江大学 | Knowledge distance-based item type learning grouping method and device |
| CN115525810A (en) * | 2022-10-08 | 2022-12-27 | 广州大学 | Context-driven dynamic grouping method for large-scale online learning |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060224532A1 (en) * | 2005-03-09 | 2006-10-05 | Case Western Reserve University | Iterative feature weighting with neural networks |
| US7792770B1 (en) * | 2007-08-24 | 2010-09-07 | Louisiana Tech Research Foundation; A Division Of Louisiana Tech University Foundation, Inc. | Method to indentify anomalous data using cascaded K-Means clustering and an ID3 decision tree |
| CN104899229A (en) * | 2014-03-07 | 2015-09-09 | 上海市玻森数据科技有限公司 | Swarm intelligence based behavior clustering system |
-
2017
- 2017-05-09 CN CN201710321172.2A patent/CN107133894B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060224532A1 (en) * | 2005-03-09 | 2006-10-05 | Case Western Reserve University | Iterative feature weighting with neural networks |
| US7792770B1 (en) * | 2007-08-24 | 2010-09-07 | Louisiana Tech Research Foundation; A Division Of Louisiana Tech University Foundation, Inc. | Method to indentify anomalous data using cascaded K-Means clustering and an ID3 decision tree |
| CN104899229A (en) * | 2014-03-07 | 2015-09-09 | 上海市玻森数据科技有限公司 | Swarm intelligence based behavior clustering system |
Non-Patent Citations (2)
| Title |
|---|
| 方平: "基于好友相似度的在线社会网络社区发现算法研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
| 马艳云等: "基于聚类和梯度选择的网络协作学习分组算法", 《电子科技》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109712011A (en) * | 2017-10-25 | 2019-05-03 | 北京京东尚科信息技术有限公司 | Community discovery method and device |
| CN109712011B (en) * | 2017-10-25 | 2022-01-07 | 北京京东尚科信息技术有限公司 | Community discovery method and device |
| CN110033239A (en) * | 2019-04-16 | 2019-07-19 | 暨南大学 | Row's test method, device, equipment and medium based on greedy algorithm |
| CN110738197A (en) * | 2019-11-12 | 2020-01-31 | 上海乂学教育科技有限公司 | Learning state evaluation model construction method |
| CN111063242A (en) * | 2020-01-14 | 2020-04-24 | 重庆工业职业技术学院 | Accounting learning method and accounting learning device |
| CN112115357A (en) * | 2020-09-11 | 2020-12-22 | 华中师范大学 | Online course forum interaction mode identification method and system |
| CN114580912A (en) * | 2022-03-07 | 2022-06-03 | 浙江大学 | Knowledge distance-based item type learning grouping method and device |
| CN114580912B (en) * | 2022-03-07 | 2024-10-18 | 浙江大学 | A project-based learning grouping method and device based on knowledge distance |
| CN115525810A (en) * | 2022-10-08 | 2022-12-27 | 广州大学 | Context-driven dynamic grouping method for large-scale online learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107133894B (en) | 2021-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107133894B (en) | Online learning grouping method based on complex network theory | |
| Gitinabard et al. | Your actions or your associates? Predicting certification and dropout in MOOCs with behavioral and social features | |
| Dunteman | Race and Sex Differences in College Science Program Participation. | |
| Dean | Resources, role models, and opportunity makers: Mentoring women in academic leadership | |
| Lawrence et al. | Implementing PBIS with fidelity: Secondary school staff members’ descriptions of school climate | |
| Singh et al. | Analysing the effect of self-efficacy and influencers on design team performance | |
| CN115525810B (en) | Context-driven large-scale online learning dynamic grouping method | |
| Lezotte | Making sense of diversity and inclusion in engineering. | |
| Luyten et al. | Dropout in the lower tracks of Dutch secondary education: Predictor variables and variation among schools | |
| Nen-Fu et al. | The Clustering Analysis System Based on Students' Motivation and Learning Behavior | |
| Paxinou et al. | Centrality Metrics from Students' Discussion Fora at Distance Education | |
| Duong et al. | Exploiting faculty evaluation forms to improve teaching quality: An analytical review | |
| CN114580912B (en) | A project-based learning grouping method and device based on knowledge distance | |
| Zhang et al. | Research on clustering mining and feature analysis of online learning behavioral data based on SPOC | |
| Madeira et al. | Prediction of student performance using rough set theory and backpropagation neural networks | |
| Luo et al. | Investigating Interactive Behaviors in Online Community of Inquiry Using Social Network Analysis | |
| Nijenhuis-Voogt et al. | Integrating CT into biology: Using decision tree models to classify cell types | |
| Shepherd et al. | War attitudes and ideological orientations of honors directors in American higher education | |
| Dong et al. | Analysis of online learning behavior based on clustering algorithm | |
| Ramasamy et al. | Analyzing Link Dynamics in Student Collaboration Networks using Canvas-A Student-Centered Learning Perspective | |
| Wilinski et al. | A method of selecting computer science students for the IT market based on their predispositions resulting from multiple intelligence | |
| Liu et al. | An Epistemic Network Analysis of Different Scripted Roles in Online Discussions | |
| Doyle | Describing and mapping the interactions between student affective factors related to persistence in science, physics, and engineering | |
| Taniguchi et al. | Self-Organizing Map Analysis of Educational Skills Using Questionnaire to University Students in Computing Classes. | |
| Ghaemi et al. | Characterizing patterns in student engagement in a team-based and project-centric course |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 510000 room 2507, 908 Tianhe North Road, Tianhe District, Guangzhou City, Guangdong Province Applicant after: Guangzhou Dayang Educational Technology Co.,Ltd. Address before: 510000 Room 3109-3113, 898 Tianbei Road, Tianhe District, Guangzhou City, Guangdong Province Applicant before: GUANGZHOU DAYANG INFO-TECH Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |