CN115545982A - Online student heterogeneous grouping system and method - Google Patents
Online student heterogeneous grouping system and method Download PDFInfo
- Publication number
- CN115545982A CN115545982A CN202211300815.2A CN202211300815A CN115545982A CN 115545982 A CN115545982 A CN 115545982A CN 202211300815 A CN202211300815 A CN 202211300815A CN 115545982 A CN115545982 A CN 115545982A
- Authority
- CN
- China
- Prior art keywords
- learning
- student
- students
- course
- online
- 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.)
- Pending
Links
Images
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—Systems or methods specially adapted for 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
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06398—Performance of employee with respect to a job function
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
Abstract
An intelligent terminal is used for online learning, an online student heterogeneous grouping system runs in a server, and the online student heterogeneous grouping system comprises a multi-mode process data acquisition module, a student portrait feature analysis module, a student heterogeneous grouping calculation module and a grouping result output module. The invention is suitable for a general online course learning platform, can realize automatic heterogeneous grouping based on the cognitive state, knowledge structure, learning emotion and learning style characteristics of students, reduces the difference between different groups in the collaborative learning task grouping in the online course learning process, promotes the collaborative complementary degree between members in the groups, fully exerts the respective advantages of the members in the groups and further improves the online collaborative learning effect.
Description
Technical Field
The invention belongs to the field of online course learning, and particularly relates to an online student heterogeneous grouping system and method.
Background
Heterogeneous grouping means that students in the same group have differences in their skills after grouping, but the differences between groups are not great in overall strength. Heterogeneous grouping is different from random grouping, i.e. students with different ability levels are artificially divided into one group, or the 'heterogeneous' is grouped according to a special requirement, so that the difference between groups is reduced, namely, the intelligent heterogeneity of the advantages in the group and the intelligent homogeneity of the advantages between the groups form a 'multi-element intelligent body' through the combination of the different intelligent properties of the advantages, and finally, the purposes of making up for the deficiencies in the group and fairly competing and justice evaluating between the groups are achieved. Research shows that members in a heterogeneous group can complement and collaborate with each other, and student members can respectively exert their own strength and learn with each other, so as to create an atmosphere of group cooperation (influence of heterogeneous group teaching in Rana. Shallow analysis university on student self-achievement and consciousness [ J ]. Modern economic information, 2016 (19): 434.); grouping intra-group heterogeneity and inter-group homogeneity, reducing inter-group difference, and improving the synchronism among the courses and the completion rate of the courses (Wangjiali, zhanghui, zhang tot. Training teaching mode of higher vocational colleges based on heterogeneous grouping is explored [ J ]. Western quality education, 2020,6 (13): 172-173.); the heterogeneous grouping cooperation learning mode can provide more language communication and practice opportunities for individual students, can fully utilize the difference among members in a group to carry out complementary teaching, and can cultivate cooperation capability; the competition among individuals is converted into the competition among groups, which is beneficial to the mind of the cooperation of the accompanying students and the consciousness of the competition; the system is beneficial to teaching according to the situation, and can make up the defect that a teacher is difficult to teach for a plurality of students with differences, thereby really achieving the goal of developing each student; in the heterogeneous grouping cooperation learning process, the difference between the teacher and each student can be acknowledged, and the potential of the difference can be fully exerted; the method is beneficial to developing the personality and meeting the requirements of students, so that the students feel confident in learning; cooperative learning activities enable students to experience mutual care and help, and enable Teachers and students to reach a boundary in harmony in the processes of multidimensional interaction, mutual whetting, getting strong and getting weak (pizzeria, exploration on heterogeneous group teaching [ J ]. Basic education forum, 2014 (25): 52-53.W.Chen, "Supporting Teachers' interpretation in collagen bundling," J.Network and Computer Applications, vol.29, pp.200-215,2006.beers, P.J., boshuizen, H.P.A., kirschner, P.A., gijselaers, W.H. (2005) Computer support for bundling in collagen bundling in Human Behavior,21 (4), 623-643.
The main factors considered when heterogeneous grouping is currently performed are the different learning styles of students (e.alfonseca, r.m.carro, E.
′n,A.Ortigosa,and P.Paredes,“The Impact of Learning Styles on Student Grouping for Collaborative Learning:A Case Study,”User Modeling and UserAdapted Interaction,vol.16,nos.3/4,pp.377-401,2006.E.
' n and P.Parees, ' Using Learning Styles for Dynamic Group Formation in Adaptive Collaborative Hypermedia Systems, ' Proc.Workshops in Connection with Fourth Int ' l Conf.Web Eng. (ICWE ' 04), pp.188-197, 2004.), where The Fisher-Sielman Learning style model is often used for The Learning style classification of students, and studies have shown that active learner-versus-eagle learner, insight learner-versus-intuitional learner-versus-learner collocation grouping can produce better Collaborative Learning effects (' Proc.IEEE's I.Rodriguez, ' The understanding Application of Learning Styles in personal visual and Collaborative Learning, 2006, ' Proc.IEEE ' l Congestion's compatible left, advanced Learning, 1142, technique 1142, pp.1142). In addition to learning style, elements such as stage test achievement, daily work achievement, prior knowledge, sex, education background, conflicting ideas or viewpoints of students are often used as important bases for heterogeneous grouping (Noroozi O, weinberger a, bi)emans H,et al.Argumentation-Based Computer Supported Collaborative Learning(ABCSCL):A synthesis of 15years of research[J]Equivalent Research Review,2012,7 (2): 79-106.) these elements tend to be considered individually or collectively when grouped, but there are few designs that are grouped from the perspective of both the student's level of cognitive mastery and knowledge structure. Meanwhile, in the specific implementation process of heterogeneous grouping, some artificial intelligence auxiliary techniques such as association rules, clustering, fuzzy genetic algorithm, fuzzy C-means, etc. (Magnisalis I, demetriadis S, karakostas A.Adaptation and Intelligent Systems for sharing Support: AReview of the Field [ J ] are adopted]IEEE Transactions on Learning Technologies,2011,4 (1): 5-20.), but still dominated by the manual participation in the analysis, developed across the face below the line.
In recent years, the development of the internet changes the production, consumption and living modes of the human society, brings a change to the social structure, and promotes the transformation of the information society to the knowledge society. The internet has become an indispensable part in our lives, online learning based on the internet becomes a new normal state, and grouping cooperation becomes an important learning form in the online learning process. Research shows that the online cooperative Learning effect of a group with high quality benefits from different special characteristics among members in the group (Rummel, n., spada, h., & Hauser, s. (2009). Learning to a color bed from describing or from observing a model. International Journal of computerized supported color Learning,26 (4), 69-92.), and is in the current state of online Learning form, if the conventional heterogeneous grouping mode is adopted, the time and labor are wasted, and the online cooperative Learning method is not practical, especially for online courses of large-scale learners, if the courses need to perform heterogeneous grouping operation of students, teachers are more attentive. Therefore, the automatic heterogeneous grouping of large-scale student groups is realized by using the past course learning data and the current course learning data of the on-line course learning of students under the condition of fully respecting the willingness and the grouping requirement of course teachers, the high-quality interaction and the improvement of the learning effect in the on-line course learning process are promoted, and the method is very urgent and has practical significance.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method has the advantages that the method is simple in process and operation, intelligent grouping result information is fed back to teachers and students in a visual chart mode, and the method is suitable for student grouping activity tasks in large-scale online course learning.
The technical scheme adopted by the invention for solving the technical problems is as follows:
an online student heterogeneous grouping system and method run in a server and comprises a multi-mode process data acquisition module, a student portrait feature analysis module, a student heterogeneous grouping calculation module and a grouping result output module, wherein:
a multimodal process data acquisition module: various modal process data generated by students in the past and when the students learn in the online course learning platform are automatically collected through a computer and stored in a multi-modal process database; the process data includes four parts: firstly, interactive behavior data comprises behavior data generated by clicking, browsing, collecting, pausing, reviewing, submitting and canceling operations; the second is to discuss the audio and video data, including the audio and video data produced in the course of online discussion, audio and video conference, online live broadcast teaching; thirdly, expression action data comprises facial expressions, emotional emotions and gesture actions expressed in the course learning process; fourthly, text publishing data comprises text data generated in the processes of commenting, posting, replying, content editing, online answering, questionnaire filling and scale filling;
student portrait feature analysis module: performing feature analysis from four dimensions of cognitive state, knowledge structure, learning emotion and learning style of students by means of four parts of data acquired in the multi-mode process data acquisition module; the cognitive state and knowledge structure dimensionality is characterized in that vector representation of the cognitive state and knowledge structure of a student is realized by utilizing interactive behavior data and text publishing data in a course learning evaluation scheme which is designed by mainly referring to the association between a course learning task and a course knowledge point established by a course teacher through a computer; the learning emotion dimension mainly realizes vector representation of eight emotions of love, pleasure, thank you, complaint, anger, disgust, fear and sadness of students by means of an expression automatic identification technology and by using discussion audio-video data and expression action data in the learning emotion dimension; the learning style dimension mainly refers to a Solomon learning style scale index system, and realizes vector representation of the learning style of students from four aspects of information processing, information perception, information input and content understanding by utilizing interactive behavior data and text publishing data in the learning style scale index system;
student heterogeneous grouping calculation module: the course teacher sets condition rules for heterogeneous grouping of students, including the number of each group and heterogeneous requirements; the heterogeneous requirements are vector dimensions participating in heterogeneous grouping calculation of students, wherein the cognitive state vector dimension and the knowledge structure vector dimension are necessary options, and the learning emotion vector dimension and the learning style vector dimension are optional items; based on four dimensional vectors of cognitive state, knowledge structure, learning emotion and learning style obtained in a student portrait feature analysis module, establishing a student grouping feature vector matrix according to heterogeneous requirements set by course teachers, automatically aggregating students into categories with the number equal to the number of people of each group by adopting a K-means algorithm, and sequentially selecting corresponding students from each category to form course learning groups with heterogeneous features among the students in the group;
a grouping result output module: and displaying the number of students, the proportion of boys and girls and the heterogeneous degree of the two students of each group in a list view mode by utilizing the student grouping information obtained by the student heterogeneous grouping calculation module, wherein the number of students is represented in a text mode, the proportion of boys and girls is represented in a bar graph mode, and the heterogeneous degree of the two students in four aspects of cognitive state, knowledge structure, learning emotion and learning style is represented in a relation graph mode.
The online student heterogeneous grouping system runs in a server, and the server can communicate with an intelligent terminal provided with an online student heterogeneous grouping system client. The intelligent terminal is a mobile phone or a tablet computer provided with an online student heterogeneous grouping system client.
The basis of the online student heterogeneous grouping system for student grouping comprises two necessary elements of cognitive state and knowledge structure and two optional elements of learning emotion and learning style.
An on-line student heterogeneous grouping method comprises the following implementation steps:
step (1) a course teacher designs an online course learning task in an online course learning platform through a computer, establishes association between the online course learning task and a course knowledge point, designs an online course learning evaluation scheme, and embeds an questionnaire developed based on a Solomon learning style scale;
the method comprises the following steps that (2) students complete online course learning tasks and fill in a Solomon learning style scale by using a mobile terminal or a computer, and meanwhile, a course teacher captures behavior interaction data, discussion audio and video data, expression action data and text publishing data generated in the online course learning process of the students through a data acquisition interface and external audio and video acquisition equipment which are embedded in the mobile terminal or the computer and stores the behavior interaction data, the discussion audio and video data, the expression action data and the text publishing data in a multi-mode process data acquisition module;
step (3) utilizing the behavior interaction data, discussion audio and video data, expression action data and text publishing data acquired in the step (2) to analyze the portrait characteristics of the student from the cognitive state, knowledge structure, learning emotion and learning style; the cognitive state is a description of the learning level of the knowledge of the students about courses, the online course learning evaluation scheme designed in the step (1) is referred, the online course learning platform automatically judges according to the score condition of the students in all learning tasks associated with the knowledge of the courses, if the score of the students in the knowledge of a certain course exceeds the passing score, the learning level of the knowledge of the course is marked as mastered, otherwise, the learning level of the knowledge of the course is marked as not mastered; the knowledge structure is characterized in that the relation acquisition level of students about course knowledge is characterized, an online course learning platform automatically judges the cognitive states of the students about the course knowledge at two ends forming a knowledge relation, if the acquisition level of the students at the two ends forming a certain knowledge relation is mastered, the acquisition level of the knowledge relation is marked as completely understood, if the course knowledge level at one end is mastered, the acquisition level is marked as partially understood, otherwise, the acquisition level is marked as completely unintelligible; the learning emotion is a description of the emotional state of the student in the online course learning process, and the automatic identification of eight expressions of love, pleasure, thank, complain, anger, disgust, fear and sadness of the student is realized by means of the current mature artificial intelligence technology or tool, so that the characterization vectors of the student on the eight expressions are formed; the learning style describes a relatively stable learning mode displayed by a student in the online course learning process, and features of the student in four aspects of information processing, perception information, input information and content understanding are sequentially mapped into a certain value on a threshold set by utilizing questionnaire information developed by the student based on a Solomon learning style scale, for example, in the aspect of information processing, if answering information of the student completely accords with an active characteristic, the value is-1, if the answering information of the student completely accords with an enthusiasm characteristic, the value is 1, otherwise, the value is another value on the threshold set;
step (4), the course teacher sets heterogeneous grouping rules, namely the number N of people in each group and the dimension participating in heterogeneous grouping calculation, wherein the cognitive state and the knowledge structure are optional dimensions, and the learning emotion and the learning style are optional dimensions;
step (5) constructing a student grouping feature vector matrix set CM by using feature information of all course students on four dimensions of cognitive state, knowledge structure, learning emotion and learning style acquired in step (3) and the heterogeneous grouping rule set in step (4); automatically aggregating the course students into N classes by using a K-means algorithm by using the CM data collected by the student grouping eigenvector matrix, sequentially selecting 1 student from the N classes of student groups to form an online learning group containing N students, and repeating the process until all students in a certain class are distributed into the group; for the rest students S in other N-1 classes, continuously adopting a K-means algorithm to gather the students into N classes, sequentially selecting 1 student from the student groups of the N classes to form an online learning group containing N students, and repeating the process until all students in a certain class are distributed into the group or when the number of the rest students in other N-1 classes meets S < N; meanwhile, in the process of sequentially selecting students from the N clusters, the difference complementation degree of the cognitive state vector and the knowledge structure vector among different students is used as an important basis for selecting heterogeneous students;
step (6), the course teacher and the students check grouping result information generated by the online student heterogeneous grouping system through the mobile terminal or the computer; the number of students, the proportion of boys and girls and the heterogeneous degree of the students in each group are presented to teachers and students in a visual graph, wherein the students can only check the information of the group in which the students are.
Preferably, the online course learning Task covered by the multimodal process data is noted as Task = { T = { T } 1 ,T 2 ,…,T t The covered course knowledge points are marked as K n ＝{K 1 ,K 2 ,…,K p N is less than or equal to p;
with student S at a certain course knowledge point K p Score of (A) is noted as
The course knowledge point K p The associated learning task is recorded as
The course knowledge point K p The associated weights occupied in the associated learning tasks are recorded in turn
Student S at course knowledge point K p The scores in the associated learning task are scored sequentially
Then
If it is
Then student S is at the point of knowledge K of the course p The learned level is marked as mastered and the value is 1, otherwise, the learned level is marked as not mastered and the value is 0, and finally, a characteristic vector cognitiveVector of the student in the cognitive state dimension is obtained and is marked as { KC 1 ,KC 2 ,…,KC p In which KC i According to the student at the course knowledge point K i The cognitive state value of (a) is 1 or 0.
Preferably, the characteristics of the students in the four aspects of information processing IP, perception information PI, input information II and content understanding CN are sequentially mapped into a set
A value of (a).
Preferably, if the student has curriculum knowledge K at both ends of a relationship of knowledge i And K j If the learning level of the knowledge relationship is mastered, the learning level of the knowledge relationship is marked as completely understood and takes a value of 1, if the knowledge level of the course at one end is mastered, the learning level is marked as partially understood and takes a value of 0, otherwise, the learning level is marked as completely unintelligible and takes a value of-1, and finally, a symmetric vector matrix of the student in the dimension of the knowledge structure is obtained
Wherein k is ij According to the student at the course knowledge point K i And K j The cognitive state value of (b) is 1, 0 or-1.
Compared with the prior art, the invention has the beneficial effects that:
(1) The invention switches in the development of the design of student grouping basis elements from two aspects of knowledge mastering level and knowledge structure of students, gives consideration to the learning emotion and learning style elements of the students comprehensively, and finally outputs grouping result information in a visual form and presents the grouping result information to course teachers and students. In the prior art, the grouping design is carried out according to a single element or external elements of students, and the internal characteristic elements such as cognitive states, knowledge structures and the like of the students are rarely concerned.
(2) According to the multivariate intelligent theory of Gardner, the invention carries out the intelligent automatic heterogeneous grouping of students on the basis of the characteristics of the learners in four aspects of cognitive state, knowledge structure, learning emotion and learning style obtained by representation and calculation, really carries out the flow design of the heterogeneous grouping of the students from the internal characteristic view of the learners, can fully exert the respective advantages of the members in the group, stimulate and promote the high-quality cooperative and mutual assistance among the members in the group to the maximum extent, avoid the homogeneity among the members in the group, ensure the fairness among the groups, save the time consumed by a large number of grouping processes, and the method has not been realized in the traditional online intelligent heterogeneous grouping of the students.
Drawings
FIG. 1 is a flow chart of an online student heterogeneous grouping system and method of the present invention;
FIG. 2 is a flow chart of an implementation of the multimodal process data acquisition module of the present invention;
FIG. 3 is a flow chart of an implementation of the student sketch feature analysis module according to the present invention;
FIG. 4 is a flow chart of the student heterogeneous group calculation module;
fig. 5 is grouping information presented to the lesson instructor and the student in the grouping result output module of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and the detailed description.
The invention relates to an online student heterogeneous grouping system and method, which comprises a multi-mode process data acquisition module, a student portrait characteristic analysis module, a student heterogeneous grouping calculation module and a grouping result output module, wherein the manufacturing process is shown in figure 1. The method comprises the following specific steps:
as shown in fig. 2, the multi-modal process data acquisition module of the present invention is specifically implemented as follows:
(1) Various modal process data generated by students in the past and in the present when learning is carried out in an online course learning platform are automatically acquired through a computer data acquisition interface and an external audio and video acquisition device termial, and an xAPI technical framework is adopted to store in a multi-modal process database in the form of information pairs < operator, action, object, operation time, result, situation and permission >, wherein the operator, the action, the object and the operation time are optional attributes, and the rest are optional attributes;
(2) The multi-modal process database mainly comprises four parts of interactive behavior data, discussion audio and video data, expression and action data and text publishing data. The interactive behavior data comprises behavior data generated by clicking, browsing, collecting, pausing, reviewing, submitting and canceling operations, and is stored in the form of < category behavior, operator, action behavior action, action object and operation time operationTime >; discussing audio and video data, including audio and video data generated in the processes of online discussion, audio and video conference and online live broadcast teaching, and storing the audio and video data in the forms of < category audio, operator participant, action, action object, operation time, duration and audio and video recording file address fileAddress >; the expression action data comprises facial expressions, emotion and gesture actions expressed in the course learning process and is stored in the forms of < category emotion, operator, action, action object, operation time operationTime and terminal > of the capture terminal; and fourthly, text publishing data, including text data generated in the processes of comment, posting, replying, editing content, answering online, filling in a questionnaire and filling in a scale, and storing in the forms of < category text, operator, action, action object, operation time operationTime and text content textualContent >.
As shown in FIG. 3, the student portrait feature analysis module of the invention is specifically implemented as follows:
(1) The course teacher designs an online course learning Task in the online course learning platform through a computer, establishing the association between the online course learning Task and the course Knowledge point Knowledge and setting the association weight omega between the two, and at a certain learning Task T i The association weight between the n lesson Knowledge points Knowlegment related with the lesson Knowledge point Knowlegment satisfies omega 1 +ω 2 +…+ω n =1; designing an online course learning evaluation scheme evaluationScheme for stipulating evaluation criteria and scoring rules of each online course learning Task and finally returning the score Grade of the student in each online course learning Task; embedding an ILS (survey questionnaire system) developed based on a Solomon learning style scale, wherein the ILS comprises 44 test questions in total, and each test question comprises two options, namely a option and b option;
(2) Performing portrait feature analysis on students from four aspects of cognitive state, knowledge structure, learning emotion and learning style by using behavior interaction data, discussion audio and video data, expression action data and text publishing data acquired from the multi-modal process data acquisition module, and recording online course learning tasks covered by the multi-modal process data as Task = { T = (T) } 1 ,T 2 ,…,T t A start-up time of the system is shortened, covered course knowledge the points are KN = { K = 1 ,K 2 ,…,K p N is less than or equal to p;
the cognitive state cognitiveState is a description of the learning level of the Knowledge of the relevant course of the student, and an online course learning evaluation scheme evaluationScheme designed in the step (1) is referred, and an online course learning platform automatically judges according to the score Performance of the student in all learning tasks Task tasks associated with the Knowledge Knowledge; student S at a certain course knowledge point K p Score of (A) is noted as
The course knowledge point K p The associated learning task is recorded as
The course knowledge point K p The associated weights occupied in the associated learning tasks are recorded in turn
Student S at course knowledge point K p The scores in the associated learning task are scored sequentially
Then
If it is used
Then student S is at the point of knowledge K of the course p The learned level is marked as mastered and the value is 1, otherwise, the learned level is marked as not mastered and the value is 0, and finally, a characteristic vector cognitiveVector of the student in the cognitive state dimension is obtained and is marked as { KC 1 ,KC 2 ,…,KC p In which KC i According to the student at the course knowledge point K i The value of the cognitive state is 1 or 0;
knowledge structure is a description of the level of acquisition of the relation between students' knowledge about courses, and an online course learning platform is used for learning knowledge K of the courses at two ends forming the relation of knowledge i And K j Automatically judging the cognitive state of the user; for all the course knowledge points associated in the same learning task, for the course knowledge points with the associated weight omega larger than a certain threshold eta, the threshold eta is set by a course teacher, and the relation exists between the course knowledge points; if the student is at both ends of course knowledge K forming a certain knowledge relationship i And K j If the learning level of the knowledge relationship is mastered, the learning level of the knowledge relationship is marked as completely understood and takes a value of 1, if the knowledge level of the course at one end is mastered, the learning level is marked as partially understood and takes a value of 0, otherwise, the learning level is marked as completely unintelligible and takes a value of-1, and finally, a symmetric vector matrix of the student in the dimension of the knowledge structure is obtained
Wherein k is ij According to the student at the course knowledge point K i And K j The value of the cognitive state is 1, 0 or-1;
learning emotion learning Emotion is a description of an emotion state of a student in an online course learning process, expression action data obtained by a multi-mode process data acquisition module is input, and automatic recognition output of eight expressions of love, pleasure, thank, complain, anger, disgust, fear and sadness of the student is realized by means of a currently mature artificial intelligence technology or tool API (application programming interface), so that a characterization vector emoteVector of the student on the eight expressions is formed and recorded as [ Elove, ejoy, egratitde, ecomint, eanger, edugust, efear and Eadness ];
the learning style is characterized by showing a relatively stable learning mode in the online course learning process of students, and the characteristics of the students in four aspects of information processing IP, perception information PI, input information II and content understanding CN are sequentially mapped into a set by utilizing questionnaire information which is filled by the students and developed based on a Solomon learning style scale
For example, in 11 questions related to the information processing IP aspect, if the answer information of the students all select the option a and completely conform to the active characteristic, the value is-1, if the answer information of the students all select the option b and completely conform to the sincere characteristic, the value is 1, otherwise, the value is a certain value in the Aggregation, and finally, the characterization vector styrevelet of the students in the learning style dimension is obtained and recorded as [ IP, PI, II, CN ] in the collection Aggregation]；
As shown in fig. 4, the heterogeneous student grouping calculation module of the present invention is specifically implemented as follows:
(1) The course teacher sets heterogeneous grouping rules, namely the number N of people of each group and the dimension participating in heterogeneous grouping calculation, wherein the cognitive state and the knowledge structure are optional dimensions, and the learning emotion and the learning style are optional dimensions;
(2) Constructing a student grouping feature vector matrix set CM by using feature information of all course students on four dimensions of a cognitive state vector cognitivive vector, a knowledge structure vector matrix, a learning emotion emotationvector and a learning style vector acquired by a student portrait feature analysis module and a heterogeneous grouping rule set in the step (1); automatically aggregating the course students into N classes by using a K-means algorithm by using the CM data collected by the student grouping eigenvector matrix, sequentially selecting 1 student from the N classes of student groups to form an online learning group containing N students, and repeating the process until all students in a certain class are distributed into the group; for the rest students S in other N-1 classes, continuously adopting a K-means algorithm to gather the students into N classes, sequentially selecting 1 student from the student groups of the N classes to form an online learning group containing N students, and repeating the process until all students in a certain class are distributed into the group or the rest students in other N-1 classes meet S < N; meanwhile, in the process of sequentially selecting students from the N clusters, the cosine similarity of the cognitive state vector and the knowledge structure vector among different students is used as an important basis for selecting the different complementary heterogeneous students, and the larger the similarity value is, the more complementary the students are.
As shown in fig. 5, the grouping result output module in the present invention is specifically implemented as follows:
<xnotran> , JSON , JSON , { { "nodes": [ { "id": idValue, "studentName": sValue }, … …, { "id": idValue, "studentName": sValue } ], "links": [ { "source": idValue, "target": idValue, "cogCos": simValue, "strCos": simValue, "emoCos": simValue, "styCos": simValue }, … …, { "source": idValue, "target": idValue, "cogCos": simValue, "strCos": simValue, "emoCos": simValue, "styCos": simValue } ] }, … …, { "nodes": [ { "id": idValue, "studentName": sValue }, … …, { "id": idValue, "studentName": sValue } ], "links": [ { "source": idValue, "target": idValue, "cogCos": simValue, "strCos": simValue, "emoCos": simValue, "styCos": simValue }, … …, { "source": idValue, "target": idValue, "cogCos": simValue, "strCos": simValue, "emoCos": simValue, "styCos": simValue } ] } }, JSON echart.js , , , , , , , , . </xnotran> Meanwhile, the course teacher can check the group information of all students through the mobile terminal or the computer, and the individual students can only check the group information of the student.
Portions of the invention not described in detail are within the skill of the art.
From the above description of the embodiments, it is clear to those skilled in the art that the present application can be implemented by software plus necessary general hardware platform. Based on this understanding, the technical solutions of the present application may be embodied in the form of software products, which essentially or partially contribute to the prior art. In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory. The computer software product may include instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the method described in the embodiments or some portions of the embodiments of the present application. The computer software product may be stored in a memory, which may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium. Computer-readable media, including both permanent and non-permanent, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information and/or information that can be accessed by a computing device. As defined herein, computer readable media does not include transitory computer readable media (transient media), such as modulated data signals and carrier waves.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the implementation of the online student heterogeneous grouping system, since the software functions executed by the processor are basically similar to those of the method implementation, the description is simple, and the relevant points can be referred to the partial description of the method implementation.
Although the present application has been described in terms of embodiments, those of ordinary skill in the art will recognize that there are numerous variations and permutations of the present application without departing from the spirit of the application, and it is intended that the appended claims encompass such variations and permutations without departing from the spirit of the application.
It should be noted that, in the description of the present application, the terms "first", "second", and the like are used for descriptive purposes only and for distinguishing similar objects, and no precedence between the two is intended or should be construed to indicate or imply relative importance. In addition, in the description of the present application, the meaning of "a plurality" is two or more unless otherwise specified.
Claims (10)
1. An online student heterogeneous grouping system, comprising: the online student heterogeneous grouping system comprises a multi-mode process data acquisition module, a student portrait feature analysis module, a student heterogeneous grouping calculation module and a grouping result output module, wherein:
a multimodal process data acquisition module: various modal process data generated by students in the past and when learning in an online course learning platform are automatically collected through a computer and stored in a multi-modal process database; the process data includes four parts: interactive behavior data, discussion audio and video data, expression action data and text publishing data; the interactive behavior data comprises behavior data generated by clicking, browsing, collecting, pausing, reviewing, submitting or canceling operations; the discussion audio and video data comprises audio and video data generated in an online discussion, audio and video conference or online live broadcast teaching process; the expression action data text publishing data comprises facial expressions, emotional emotions or gesture actions expressed in the course learning process; the text publishing data comprises text data generated in the processes of comment, posting, replying, content editing, online answering, questionnaire filling or scale filling;
student portrait feature analysis module: performing feature analysis from four dimensions of cognitive state, knowledge structure, learning emotion and learning style of students by means of four parts of data acquired in the multi-mode process data acquisition module; the cognitive state and knowledge structure dimensionality is characterized in that the vector representation of the cognitive state and knowledge structure of the student is realized by utilizing interactive behavior data and text publishing data in a course learning evaluation scheme which is designed by mainly referring to the association between a course learning task and a course knowledge point established by a course teacher through a computer; the learning emotion dimensionality is characterized in that vector representation of eight emotions, namely love, pleasure, thank you, complain, anger, disgust, fear and sadness, of a student is realized by mainly utilizing discussion audio-video data and expression action data in an expression automatic identification technology; the learning style dimension mainly refers to a Solomon learning style scale index system, and realizes vector representation of the learning style of students from four aspects of information processing, information perception, information input and content understanding by utilizing interactive behavior data and text publishing data in the learning style scale index system;
student heterogeneous grouping calculation module: the course teacher sets condition rules for heterogeneous grouping of students, including the number of each group and heterogeneous requirements; the heterogeneous requirements are vector dimensions participating in heterogeneous grouping calculation of students, wherein the cognitive state vector dimension and the knowledge structure vector dimension are indispensable items, and the learning emotion vector dimension and the learning style vector dimension are optional items; based on four dimensional vectors of cognitive state, knowledge structure, learning emotion and learning style obtained in a student portrait feature analysis module, establishing a student grouping feature vector matrix according to heterogeneous requirements set by a course teacher, automatically aggregating students into class numbers with the number equal to the number of people of each group by adopting a K-means algorithm, and sequentially selecting corresponding students from each class to form course learning groups with heterogeneous features among the students in the group;
a grouping result output module: and displaying the number of students, the proportion of boys and girls and the heterogeneous degree of the two students of each group in a list view mode by utilizing the student grouping information obtained by the student heterogeneous grouping calculation module, wherein the number of students is represented in a text mode, the proportion of boys and girls is represented in a bar graph mode, and the heterogeneous degree of the two students in four aspects of cognitive state, knowledge structure, learning emotion and learning style is represented in a relation graph mode.
2. The online student heterogeneous grouping system of claim 1 wherein: the online student heterogeneous grouping system runs in a server, and the server can communicate with an intelligent terminal provided with an online student heterogeneous grouping system client.
3. The online student heterogeneous grouping system according to claim 1, wherein: the basis of the online student heterogeneous grouping system for student grouping comprises two necessary elements of cognitive state and knowledge structure and two optional elements of learning emotion and learning style.
4. An on-line student heterogeneous grouping method is characterized by comprising the following implementation steps:
step (1), a course teacher designs an online course learning task in an online course learning platform through a computer, establishes association between the online course learning task and a course knowledge point, designs an online course learning evaluation scheme, and embeds an investigation questionnaire developed based on a Solomon learning style scale;
step (2), the students complete online course learning tasks and fill in a Solomon learning style table by using the mobile terminals or the computers, and meanwhile, the course teachers capture behavior interaction data, discussion audio and video data, expression action data and text publishing data generated in the online course learning process of the students through data acquisition interfaces and external audio and video acquisition equipment which are embedded in the mobile terminals or the computers and store the data in a multi-mode process data acquisition module;
step (3), performing portrait feature analysis on the student from four aspects of cognitive state, knowledge structure, learning emotion and learning style by using the behavior interaction data, discussion audio and video data, expression action data and text publishing data acquired in the step (2); the cognitive state is characterized in that the learning level of knowledge of students about courses is portrayed, according to the online course learning evaluation scheme designed in the step (1), an online course learning platform automatically judges the learning level of the relationship between the knowledge of students about courses according to the scoring condition of the students in all learning tasks related to the course knowledge, the online course learning platform automatically judges the cognitive state of the knowledge of the courses at two ends forming the knowledge relationship according to the students, the learning emotion is the portrayal of the emotional state of the students in the online course learning process, and the automatic recognition of eight expressions of love, happiness, thank you, complaint, anger, nausea, fear and sadness of the students is realized by means of the current mature artificial intelligence technology or tool to form the characterization vectors of the students on the eight expressions; the learning style describes a relatively stable learning mode displayed in the online course learning process of the student, and features of the student in four aspects of information processing, perception information, input information and content understanding are sequentially mapped to a certain value on a threshold set by utilizing questionnaire information which is filled by the student and developed based on a Solomon learning style scale;
step (4), the course teachers set heterogeneous grouping rules, namely the number N of people in each group and the dimensionality participating in heterogeneous grouping calculation, wherein the cognitive state and the knowledge structure are selected dimensionalities, and the learning emotion and the learning style are selected dimensionalities;
step (5), utilizing the feature information of all course students in the cognitive state, knowledge structure, learning emotion and learning style obtained in the step (3) and the heterogeneous grouping rules set in the step (4),
constructing a student grouping feature vector matrix set CM;
automatically aggregating the course students into N classes by using a K-means algorithm by using the CM data of the student grouping eigenvector matrix set, sequentially selecting 1 student from the N classes of student groups to form an online learning group containing N students, and repeating the process until all students in a certain class are distributed into the group; for the rest students S in other N-1 classes, continuously adopting a K-means algorithm to gather the students into N classes, sequentially selecting 1 student from the student groups of the N classes to form an online learning group containing N students, and repeating the process until all students in a certain class are distributed into the group or the rest students in other N-1 classes meet S < N; meanwhile, in the process of sequentially selecting students from the N clusters, the difference complementation degree of the cognitive state vector and the knowledge structure vector among different students is used as an important basis for selecting heterogeneous students;
step (6), the course teachers and the students check grouping result information generated by the online student heterogeneous grouping system through the mobile terminals or the computers; the number of students, the proportion of boys and girls and the heterogeneous degree of the students in each group are presented to teachers and students in a visual graph, wherein the students can only check the information of the group in which the students are.
5. The online student heterogeneous grouping method according to claim 4 wherein in the automatic evaluation of the lesson knowledge acquisition level characterization, if a student scores over a passing score for a lesson knowledge, the lesson knowledge acquisition level is marked as mastery, otherwise the lesson knowledge acquisition level is marked as not mastery.
6. The online student heterogeneous grouping method according to claim 4, wherein in the automatic evaluation of the relationship acquisition level delineation between the course knowledge, if the acquisition level of the course knowledge of the student at both ends constituting a certain knowledge relationship is mastered, the acquisition level of the knowledge relationship is marked as completely understood, if only the course knowledge level at a certain end is mastered, the acquisition level is marked as partially understood, otherwise the acquisition level is marked as completely unintelligible.
7. The online heterogeneous student grouping method according to claim 4, wherein in the step (3), in the aspect of information processing, the value is-1 if the response information of the student completely conforms to the active type feature, the value is 1 if the response information of the student completely conforms to the meditation type feature, and the value is other value on the threshold set.
8. The online student heterogeneous grouping method of claim 4 wherein the online course learning tasks covered by the multi-modal process data are denoted as Task = { T = { T } 1 ,T 2 ,…,T t The covered course knowledge points are marked as K n ＝{K 1 ,K 2 ,…,K p N is less than or equal to p;
with student S at a certain course knowledge point K p Score of (1) is noted as
The course knowledge point K p The associated learning task is noted
The course knowledge point K p The associated weights occupied in the associated learning tasks are recorded in turn
Student S at course knowledge point K p The scores in the associated learning tasks are scored sequentially
Then the
If it is
Then student S is at this curriculum knowledge point K p The learned level is marked as mastered and the value is 1, otherwise, the learned level is marked as not mastered and the value is 0, and finally, a characteristic vector cognitiveVector of the student in the cognitive state dimension is obtained and is marked as { KC 1 ,KC 2 ,…,KC p In which KC i According to the student at the course knowledge point K i The cognitive state value of (a) is 1 or 0.
9. The online student heterogeneous grouping method according to claim 4, wherein the features of students in the four aspects of information processing IP, perception information PI, input information II and content understanding CN are mapped to a set in sequence
A value of (a).
10. An on-line student heterogeneous grouping method as claimed in claim 8 wherein if students have course knowledge K at both ends of a knowledge relationship i And K j If the learning level of the course is mastered, the learning level of the knowledge relationship is marked as completely understood and 1, and if the learning level of the course at one end is mastered, the learning level of the course at the other end is marked as completely understood and 1Partially understanding, namely taking the value as 0, otherwise marking the value as completely unintelligible, taking the value as-1, and finally obtaining the symmetric vector matrix of the student on the knowledge structure dimension
Wherein k is ij According to the student at the course knowledge point K i And K j The cognitive state value of (b) is 1, 0 or-1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211300815.2A CN115545982A (en) | 2022-10-24 | 2022-10-24 | Online student heterogeneous grouping system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211300815.2A CN115545982A (en) | 2022-10-24 | 2022-10-24 | Online student heterogeneous grouping system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115545982A true CN115545982A (en) | 2022-12-30 |
Family
ID=84718913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211300815.2A Pending CN115545982A (en) | 2022-10-24 | 2022-10-24 | Online student heterogeneous grouping system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115545982A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117670620A (en) * | 2024-01-31 | 2024-03-08 | 深圳市康莱米电子股份有限公司 | Education flat-panel intelligent interaction method, system and equipment |
| CN117670620B (en) * | 2024-01-31 | 2024-05-14 | 深圳市康莱米电子股份有限公司 | Education flat-panel intelligent interaction method, system and equipment |
-
2022
- 2022-10-24 CN CN202211300815.2A patent/CN115545982A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117670620A (en) * | 2024-01-31 | 2024-03-08 | 深圳市康莱米电子股份有限公司 | Education flat-panel intelligent interaction method, system and equipment |
| CN117670620B (en) * | 2024-01-31 | 2024-05-14 | 深圳市康莱米电子股份有限公司 | Education flat-panel intelligent interaction method, system and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Somech et al. | Organizational citizenship behavior in schools: Examining the impact and opportunities within educational systems | |
| Easterby-Smith et al. | Constructing contributions to organizational learning: Argyris and the next generation | |
| Jarvis | Adult and continuing education: Theory and practice | |
| Lian et al. | Rhizomatic learning:“As… when… and if…” A strategy for the ASEAN community in the 21st century | |
| CN109902912B (en) | Personalized image aesthetic evaluation method based on character features | |
| CN110782375B (en) | Online learning overall process dynamic analysis method and system based on data | |
| Johnson | Teaching strategies for all teachers: Enhancing the most significant variable | |
| Shan et al. | Blended teaching design of college students’ mental health education course based on artificial intelligence flipped class | |
| Pell et al. | Promoting group work at key stage 3: Solving an attitudinal crisis among young adolescents? | |
| Howe | Strategies for supporting the transition from small-group activity to student learning: A possible role for beyond-group sharing | |
| Gupta et al. | Technological disruptions reshaping management education: a qualitative study of stakeholder perspectives | |
| Wambsganss et al. | Designing a conversational agent as a formative course evaluation tool | |
| Chen et al. | Integration of Big Data and Artificial Intelligence in Constructing Learners' Individualized Feedback System | |
| Salem et al. | Establishment of smart education system in modern universities: Concept, technologies and challenges | |
| CN115545982A (en) | Online student heterogeneous grouping system and method | |
| CN109299805A (en) | A kind of online education course request processing method based on artificial intelligence | |
| PEROVIĆ | Data mining influence on e-learning | |
| Sun et al. | Study on the distance learners’ academic emotions using online learning behavior data | |
| Zhang et al. | Construction and research of constructive English teaching model applying multimodal neural network algorithm | |
| Luo et al. | Investigating Interactive Behaviors in Online Community of Inquiry Using Social Network Analysis | |
| Hamdun | The Role of Parenting Styles in Internalizing Islamic Moderation Values in Children: A Phenomenological Study | |
| Liu et al. | The Design of a Moral Education Evaluation System for College Students Based on a Deep Learning Model | |
| Macauley | Pedagogic continuity in doctoral supervision: passing on, or passing by, of information skills | |
| Arunachalam et al. | Measures for predicting success factors of ELearning in Educational institutions | |
| Sękala et al. | Technical Education for Industry 4.0: Generation Z's motivation to study technical subjects-a comparative study of selected universities in Denmark and Poland |
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 |