CN113378076B - Online education-oriented learner collaborative learning social relationship construction method - Google Patents
Online education-oriented learner collaborative learning social relationship construction method Download PDFInfo
- Publication number
- CN113378076B CN113378076B CN202110730170.5A CN202110730170A CN113378076B CN 113378076 B CN113378076 B CN 113378076B CN 202110730170 A CN202110730170 A CN 202110730170A CN 113378076 B CN113378076 B CN 113378076B
- Authority
- CN
- China
- Prior art keywords
- learner
- learning
- behavior
- basic
- partner
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000010276 construction Methods 0.000 title claims abstract description 13
- 230000006399 behavior Effects 0.000 claims abstract description 94
- 238000013528 artificial neural network Methods 0.000 claims abstract description 36
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 230000000694 effects Effects 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000012300 Sequence Analysis Methods 0.000 claims abstract description 8
- 238000003064 k means clustering Methods 0.000 claims abstract description 8
- 238000012549 training Methods 0.000 claims abstract description 6
- 230000002452 interceptive effect Effects 0.000 claims description 31
- 230000003993 interaction Effects 0.000 claims description 26
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000009826 distribution Methods 0.000 claims description 9
- 238000002372 labelling Methods 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 7
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 4
- 238000011002 quantification Methods 0.000 claims description 4
- 230000001502 supplementing effect Effects 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 3
- 230000008451 emotion Effects 0.000 claims description 3
- 238000011049 filling Methods 0.000 claims description 3
- 238000010438 heat treatment Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 3
- 230000006641 stabilisation Effects 0.000 claims description 3
- 238000011105 stabilization Methods 0.000 claims description 3
- 239000013589 supplement Substances 0.000 claims 1
- 230000008569 process Effects 0.000 description 11
- 238000004891 communication Methods 0.000 description 8
- 239000013598 vector Substances 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000001447 compensatory effect Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000002996 emotional effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000000454 anti-cipatory effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000010413 gardening Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 235000012054 meals Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 238000003012 network analysis Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
- 230000002747 voluntary effect Effects 0.000 description 1
- 238000010792 warming Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
- G06Q50/01—Social networking
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Databases & Information Systems (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Educational Technology (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention discloses a construction method of a learner collaborative learning social relationship facing online education, which comprises the following steps: acquiring basic information data of a learner in an online education platform, and extracting attribute characteristics of the basic information data by adopting a sequence analysis method; performing behavior analysis on the basic behavior characteristics in the attribute characteristics to label the partnership; processing the partnership labels by using a k-Means clustering algorithm to obtain the learning community clustering discovery of the learner; constructing a general neural network according to the learning community clustering discovery of the learner, training the collaborative learning social relationship neural network of the learner by using the attribute characteristics to obtain the collaborative learning social relationship neural network of the learner, and further obtaining a current buddy recommendation list; and (3) converging the learner partner network and rearranging the partner recommendation list based on the scale-free network theory and the social triangle theory. The method can recommend a proper learning partner for the learner, and further enhance the learning effect of the learner on the platform.
Description
Technical Field
The invention relates to the technical field of online education platforms, in particular to a construction method of a learner collaborative learning social relationship for online education.
Background
With the rapid development of information technology, the production, life and working modes of people are changing, and the world of our life is increasingly in the aspects of internet of things, interconnection and intellectualization. The educational objectives and teaching modes under the background of the traditional industrial era are no longer suitable for the requirements of the modern informatization and big data era. Various institutions, researchers and the like provide a lot of new innovation points for the implementation mode of intelligent education, such as providing application practice forms of an online classroom and a turnover classroom, and changing the previous teaching mode, so that learners can obtain more autonomous learning time and corresponding learning resources, and the learning process is more efficient. With the proposition and the gradual popularization of novel education mode Online education (e-Learning) and MOOC (Massive Open Online Courses), more learners have the opportunity to visit and learn on an Online platform, and accordingly, the platform accumulates Massive teaching behavior data and knowledge resources and provides a good foundation for the updating and perfection of the platform. However, the online education provides more abundant learning resources and simultaneously limits the learning platform of learners to the network compared with the conventional school education, and the learning mode of mutual communication between learners, which is very important in the learning life of the conventional school, is split due to the isolation of learners from each other in physical space. Each learner basically learns independently in the learning process. However, in the case of a course with a great difficulty, the learner cannot solve the confusion in an effective communication manner, which greatly affects the learning effect and learning enthusiasm of the learner on the online education platform, and these problems usually cause the learner to have learning frustration that is disoriented and is rather useless. In order to enable learners to better understand the learning content of the online education platform, improve the learning efficiency, communicate with other learners more fully, establish good partnership, and improve the platform use experience and enthusiasm of the learners, innovations need to be made on the method for providing learning partners for the learners of the online education platform.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, the invention aims to provide a construction method of a learner collaborative learning social relationship facing online education, which can recommend a proper learning partner for the learner, and further enhance the learning effect of the learner on a platform.
In order to achieve the above purpose, an embodiment of the present invention provides a method for constructing a learner collaborative learning social relationship for online education, including the following steps: step S1, obtaining basic information data of a learner in an online education platform, and extracting attribute characteristics of the basic information data by adopting a sequence analysis method, wherein the attribute characteristics comprise classified learner behavior characteristics, basic distances among various subjects and interactive positive data; s2, performing behavior analysis on the basic behavior characteristics in the attribute characteristics to label the partnership; s3, processing the partnership label by using the k-Means clustering algorithm to obtain a learning community clustering discovery of the learner; s4, constructing a general neural network according to the learner learning community clustering discovery, and training the learner collaborative learning social relationship neural network by using the attribute characteristics to obtain the learner collaborative learning social relationship neural network so as to obtain a current buddy recommendation list; and S5, basically sequencing the partners of the current learner in the converged state in the learner co-learning social relation neural network based on a scale-free network theory and a social triangle theory, and supplementing the current partner recommendation list according to a recommendation system long-tail theory.
In addition, the construction method for the learner to collaboratively learn the social relationship for online education according to the above embodiment of the present invention may further have the following additional technical features:
further, in an embodiment of the present invention, the step S1 specifically includes: step S101, acquiring initial information data of a learner in an online education platform, and performing null filling and basic unique heat treatment on the basic information data to obtain basic information data, wherein the basic information data comprises a behavior sequence, a community interaction behavior record, subject preferences, learning time, a learning place and a system login mode; step S102, extracting the basic information data by adopting a sequence analysis method to obtain basic behavior characteristics and a corresponding vectorization result; step S103, processing the basic behavior characteristics, the corresponding vectorization results and the community interaction behavior records according to a psychological personality dividing and scoring mechanism to obtain a personality characteristic value of the learner; step S104, carrying out weighted average numerical calculation on the basic behavior characteristics and the learner character characteristic numerical values to obtain a primary numerical value, calculating the average social activity and platform activity of a user as the learner behavior characteristics, and carrying out Gaussian distribution and classification on the learner behavior characteristics according to the characteristics of Gaussian distribution of behavior attributes in the crowd to obtain classified learner behavior characteristics; step S105, carrying out quantitative processing on the learner subject preference in the basic information data by adopting word2vec to obtain the basic distance between the subjects; and step S106, performing word segmentation analysis on the community interaction behavior record in the basic information data by adopting a word segmentation tool, and performing numerical quantification processing on whether the current interaction is positive according to Chinese word emotion values to obtain interaction positive data.
Further, in an embodiment of the present invention, the step S2 specifically includes: step S201, dividing and filtering the basic behavior characteristics in the attribute characteristics to obtain the real interactive behavior of the learner and the non-interactive behavior of the learner; step S202, analyzing the real interactive behavior of the learner and the no interactive behavior of the learner to derive the partnership label.
Further, in an embodiment of the present invention, the step S3 specifically includes: step S301, defining the learner partner interactive behavior in the partnership labeling; step S302, calculating the Jaccard Distance between learners according to the defined learner partner interactive behaviors; and step S303, processing the Jaccard Distance between learners by using the k-Means clustering algorithm to obtain the learning community clustering discovery of learners.
Further, in an embodiment of the present invention, the step S4 specifically includes: step S401, obtaining a time series relationship through learner learning community clustering discovery, dividing a learner behavior sequence according to the time series relationship, and obtaining a recommendation buddy list through calculation of a learner sequence with later time and a learner sequence with earlier time; step S402, interactive behaviors between the learner and other users are found and obtained through the learner learning community clustering so as to label the partnership between the learner and the other users; and S403, constructing a global neural network according to the recommendation partner list and the labeled partnership of the recommendation partner and the labeled recommendation partner based on deep learning, training the global neural network by using the attribute characteristics to obtain a learner collaborative learning social relationship neural network, and further obtaining a current partner recommendation list.
Further, in an embodiment of the present invention, the step S5 specifically includes: step S501, examining the current social network behavior in the learner co-learning social relationship neural network, determining whether the basic scale-free network characteristics and the evolution mode are met, and if so, executing step S502; step S502, evolution of the current partner network is carried out according to the probability connection and enhancement mode of the scale-free network until the convergence state is reached; and S503, based on the social triangle stabilization principle, basically sequencing the buddies of the current learner according to the buddies of the current learner in the convergence state, and supplementing the current buddy recommendation list according to the long-tailed theory of the recommendation system.
Further, in an embodiment of the present invention, the method further includes: and S6, tracking basic information data of the learner to periodically update the characteristic attribute and the partner network structure of the learner, and performing new learner recommended partner relation calculation according to the new personalized characteristic attribute of the learner.
The construction method for the learner to learn the social relationship in a collaborative manner facing the online education, provided by the embodiment of the invention, has at least the following advantages:
(1) Most learners of the existing online education platform watch videos to learn, and hardly participate in a discussion area to carry out valuable discussion learning, the embodiment of the invention makes up the dilemma that learners are difficult to carry out interactive learning under the conditions of numerous participation numbers, anonymous identities and non-contact of online education learners, emphasizes the matching property of the learners in the recommendation process of learning partners and the purpose of improving the learning effect and efficiency, ensures that the learning process of the learners is not independently learnt by one person any more, and is easier and more pleasant;
(2) The embodiment of the invention emphasizes that by means of basic inherent attributes of the learner, such as age, gender, academic history and the like, and accumulated behavior characteristics and character characteristic data, a personalized model is established for the learner, a learning partner which is most suitable for the aspects of personal habits, characters and the like of the learner can be screened out for the learner, personalized learning partner recommendation is provided, the learner can answer questions and solve confusion in the communication with other users, the learning efficiency is improved, the learning process is not boring and monotonous, all problems can be solved in time through the communication, and the 'bonbonboned meal' is not eaten;
(3) The embodiment of the invention not only comprises the extraction of the characteristic vector and the algorithm design of the learner, but also comprises the system design of the interaction platform of the learner, and the recommendation partners of the online education platform and the friends of the interaction platform are directly mapped, thereby better providing services for the learner;
(4) According to the embodiment of the invention, the behavior data of the platform user is continuously collected, the personalized feature data corresponding to the learner is synchronously updated, and the personalized feature data is dynamically updated into the learning partner list recommended by the learner, so that the personalized feature data can be updated along with time, the learner can be continuously depicted in a refined manner, and the recommendation pertinence of the learning partners is stronger and stronger.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a flow chart of a construction method of a learner collaborative learning social relationship for online education according to an embodiment of the present invention;
FIG. 2 is a diagram of learner trait attributes in accordance with one embodiment of the present invention;
FIG. 3 is a partnership algorithm annotation for an online education platform in accordance with the present invention;
FIG. 4 is a flowchart illustrating the detailed implementation of step S3 according to an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of the neural network for learner collaborative learning social relationship constructed in step S4 according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of an interactive network simulation convergence algorithm for learner recommendation of buddy lists according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a converged network partner reordering algorithm implementation in accordance with an embodiment of the present invention;
FIG. 8 is a diagram of an online education platform learner interaction system architecture in accordance with one embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
First, in the learning process of the online education platform, the course generally includes several links of basic video learning, courseware examination, post-school learning, discussion in discussion area, final major homework and end-of-term examination. By recording action categories, duration and the like of the learner on the platform, a behavior sequence of the learner on the current platform with a time tag can be constructed. However, these data are often only collected, and personalized attributes such as learning status, ability, and features of the learner cannot be analyzed, and thus, the data are not fully utilized. The embodiment of the invention provides the full utilization of a large amount of collected learner behavior data and basic characteristics, and particularly provides the analysis of the behavior sequence and the basic attributes of a learner, so that an adaptive learning partner is recommended for the learner to enhance the learning effect on a platform, and the trouble caused by individual learning is reduced. The method for constructing the learner collaborative learning social relationship for online education according to the embodiment of the present invention is specifically described below.
The following describes a construction method of a learner collaborative learning social relationship for online education proposed according to an embodiment of the present invention with reference to the accompanying drawings.
Fig. 1 is a flowchart of a method for constructing a learner collaborative learning social relationship for online education according to an embodiment of the present invention.
As shown in fig. 1, the construction method of the learner collaborative learning social relationship facing the online education comprises the following steps:
in step S1, basic information data of a learner in the online education platform are obtained, and attribute characteristics of the basic information data are extracted by adopting a sequence analysis method, wherein the attribute characteristics comprise classified learner behavior characteristics, basic distances among disciplines and interactive positive data.
That is, step S1 performs feature attribute extraction by analyzing the sequence of learner' S behaviors.
It should be noted that, before step S1, the learner is required to perform basic attribute completion based on personal interests on the online education platform, and then the online education platform performs filtering on the learner of the course, and the learner who simply knows or browses the course content will be filtered without performing deep analysis and partner recommendation, i.e. not being the basic information data of the learner in step S1.
Further, step S1 specifically includes:
step S101, obtaining initial information data of a learner in an online education platform, and performing null filling and basic unique heat treatment on the basic information data to obtain basic information data, wherein the basic information data comprises a behavior sequence, community interaction behavior records, subject preferences, learning time, a learning place and a system login mode.
It should be noted that the corresponding behavior sequence of the learner on the current platform is generated according to the time sequence through the browsing record, the learning record, the problem record, the discussion record of the discussion area and the like of the corresponding learner in the database (i.e. the initial information data).
And step S102, extracting the basic information data by adopting a sequence analysis method to obtain basic behavior characteristics and a corresponding vectorization result.
That is to say, the basic information data is extracted by adopting a sequence analysis mode, and the basic behavior characteristics and the corresponding vectorization results of the learner in the aspects of the common learning time, the learning place and the common learning system are obtained.
And step S103, processing the basic behavior characteristics, the corresponding vectorization results and the community interaction behavior records according to a psychological personality dividing and scoring mechanism to obtain the personality characteristic values of the learners, and meanwhile, carrying out database localized storage.
And step S104, carrying out weighted average numerical calculation on the basic behavior characteristics and the learner character characteristic numerical values to obtain a primary numerical value, taking the average social activity and the platform activity of the user as the behavior characteristics of the learner, and carrying out Gaussian distribution and classification on the behavior characteristics of the learner according to the characteristics of Gaussian distribution of behavior attributes in the crowd to obtain the classified behavior characteristics of the learner.
Specifically, the basic behavior characteristics of the learner, including the video watching rate, the video watching speed, the average watching times of the watched videos, the average job completion rate, the average post number and the average praise number, are subjected to weighted average numerical calculation by adopting a numerical weighting mode to obtain a primary numerical value, and secondary numerical values, such as the average social activity of a user, the platform activity and the like, are calculated to serve as the behavior characteristics of the learner, and the behavior characteristics of the learner are subjected to Gaussian distribution and classification according to the characteristics of Gaussian distribution of behavior attributes in the crowd to obtain the classified behavior characteristics of the learner.
Step S105, carrying out quantitative processing on the learner discipline preference in the basic information data by adopting word2vec to obtain the basic distance between the disciplines.
And S106, performing word segmentation analysis on the community interaction behavior record in the basic information data by adopting a word segmentation tool, and performing numerical quantification processing on whether the current interaction is positive according to Chinese word emotion values to obtain interaction positive data.
For example, as shown in fig. 2, since the learner cannot directly communicate with the partner face to face in the conventional teaching environment, the embodiment of the present invention proposes the partner recommended activities of the learner, and the extracted contents and the corresponding calculation method can be as follows:
the embodiment of the invention combines basic contents of educational psychology to divide attributes of learners to be investigated into the following types:
similarity factors: students have high similarity in terms of belief, attitude, value, identity, cultural degree and social background, and are easy to approach each other and form a positive interpersonal relationship, and students have low similarity and are easy to be remote or even generate a negative interpersonal relationship. The basic attribute and character feature (operation data obtained by other algorithms) of the learner are the intrinsic characteristic attribute of the learner, and the basic attribute and the character feature do not change too much in a short time. And (4) carrying out single-hot coding treatment after marking by adopting a classification mode.
Compensatory factors: the compensatory factor is a relatively important factor in that interpersonal relationships are constrained by anticipatory compensation for each other. Students are eager to achieve certain purposes in communication and interpersonal relationship and meet certain needs, and the voluntary or involuntary social motivation is a natural psychological tendency of individuals. Such compensation includes obtaining others affirmation, sympathy, reluctance, support and help, etc. On an online education platform, the recognition and the consistency of other learners on the learning habits and the learning intentions of the learners are important partnership construction elements. Therefore, the embodiment of the invention takes out the subject preference attribute corresponding to the learner and the corresponding course learning intention attribute as the characteristic extraction of the compensatory factors. Wherein the subject preference is obtained according to course selection tendency, and the preference has 15 selectable options of computer, foreign language, management, philosophy, economics, law, education and teaching, literature and culture, history, science, engineering, agriculture, forestry and gardening, medicine and health, art design and the like. The corresponding learning intention is the basic result data obtained by other algorithms and is directly called for use. Both course level intents and chapter level intents. The course level intentions include trial listening to courses, browsing reference, missing and missing checking, score obtaining and certificate obtaining, and the chapter level intentions include browsing reference, learning knowledge and score obtaining, so that the intentions are strengthened step by step.
The learner's behavior characteristics include two parts: basic learning habits of learners and historical interaction engagement of learners. The basic learning habits are divided into different action categories, duration and the like through the basic behavior sequence of the learner constructed before, weighted post-processing is carried out to obtain specific numerical values, then normal distribution standardization is carried out, and the groups with different habits are divided through quantiles on the right sides of 0.3 and 0.1. And the historical interaction integrating degree of the learner is processed aiming at the interaction record between the currently examined learner and a certain platform user. The former discussion content adopts the ending word segmentation and then carries out overall emotional scoring through the Chinese word emotional dictionary to judge whether the current user and the learner have quarreling before or have good conversation, and then obtains the corresponding interactive positive data or the fitness score.
In step S2, performing behavior analysis on the basic behavior features in the attribute features to obtain a partnership label.
In other words, the learner's interactive behavior in the online education platform is analyzed for partnership labeling.
Further, step S2 specifically includes:
step S201, dividing and filtering basic behavior characteristics in attribute characteristics to obtain real interactive behaviors of learners and non-interactive behaviors of learners;
and step S202, analyzing the real interactive behaviors of the learner and the non-interactive behaviors of the learner to derive the partnership label.
Specifically, as shown in fig. 3, in a conventional educational environment, the inter-student partnership labeling is generally performed by means of questionnaires, direct inquiry, indirect observation, or the like. However, unlike the conventional school education environment, the communication between learners of the online education platform is too rare, and through data statistics, learners actively participating in discussion in the discussion area are less than 1% of the whole learners and the content in the aspect of effectively discussing and learning is less than 2%. Therefore, the embodiment of the invention calls the attribute characteristics obtained in the step S1 to carry out vector distance calculation, then uses the basic exponential function to carry out exponential operation on each distance value, and carries out MIN-MAX normalization processing on the optional partner set of each user. Due to the lack of direct interactive content, logic errors are judged after direct sequencing, and therefore the embodiment of the invention adopts a random number judgment mode to label corresponding partnership. It can be appreciated that learners with closer vector distances are more likely to be determined to be learning partners. For a few learner pairs with direct communication, whether the social contact degree obtained in step S1 is positive conversation is determined, and phenomena such as quarrel, 35881, and curse do not exist. If not, the learning partner is marked.
In step S3, the partnership labeling is processed by using a k-Means clustering algorithm, and the learning community clustering discovery of the learner is obtained.
That is, a basic partner network, namely learner learning community cluster discovery, is constructed based on a k-Means clustering algorithm.
Further, in an embodiment of the present invention, step S3 specifically includes:
step S301, defining the learner partner interaction behavior in the partnership labeling;
step S302, calculating the Jaccard Distance between learners according to the defined learner partner interactive behaviors;
step S303, the Jaccard Distance between learners is processed by utilizing a k-Means clustering algorithm to obtain the learning community clustering discovery of learners.
Specifically, interactive behaviors between learners include the following: learning a course together for more than a period of time, discussing in a forum, completing a project or major work together, from the same school in the background and similar in age. If one or more connections exist between the learner and a user, an interactive edge is constructed between two persons, the value is 1, and the method is different from the traditional vector clustering method based on k-Means. Because various factors cannot be completely determined on the weight of the vector distance influence in education, the subjective human factor influence in community discovery can be more effectively solved by adopting the Jaccard distance to judge the distance between different learners. After the community classification is completed, users in the same community can be used as optional recommendation learning partners to enter the next step of neural network more refined calculation and recommendation.
In step S4, a general neural network is constructed according to the learners ' learning community clustering discovery, and the learners ' collaborative learning social relationship neural network is trained by using the attribute characteristics to obtain the learners ' collaborative learning social relationship neural network, so as to obtain the current buddy recommendation list. Namely, constructing and training an overall neural network based on learners' learning of community cluster discovery.
Further, in an embodiment of the present invention, step S4 specifically includes:
step S401, obtaining a time series relationship through learner learning community clustering discovery, dividing a learner behavior sequence according to the time series relationship, and obtaining a recommendation buddy list through calculation of a learner sequence with later time and a learner sequence with earlier time;
step S402, interactive behaviors between the learner and other users are obtained through learner learning community clustering discovery so as to label the partnership of the learner and the other users;
step S403, based on methods such as deep learning LSTM and embedding, a general neural network is constructed according to the recommendation buddy list and the labeled buddy relationship, the general neural network is trained by using the attribute characteristics, a social relationship neural network for the learner to learn cooperatively is obtained, and then the current buddy recommendation list is obtained.
Specifically, as shown in fig. 5, the input end of the learner collaborative learning social relationship neural network includes user-based data, which is formed according to the learner's learning habits extracted from the learner's past learning experience, the learner's subject preferences, and the learner's learning intention on the course. And the output end is the classification result of 0 or 1, namely whether the user to be examined currently can become a good learning partner of the learner in the course. The design of the neural network respectively adopts the following basic structures from front to back:
1) One-hot coding layer:
most basic information and learning habits of learners are classified into discrete categories. And by adopting one-hot mode coding, the data can be converted into a mode with clear meaning and easy calculation by a machine learning neural network.
2) Embedding/Autoencoder layer:
in the current neural network operation process, the dimension of an input vector is basically required to be an integral power or a combination of 2. It is clear that many neural network input features extracted by the present subject are not an integer power of 2 after the one-warming. Therefore, it is compressed into the closest vector of power of 2 dimension using autoencoder's approach. The calculation is more convenient and faster, and meanwhile, the influence of human factors brought by labeling can be reduced under the condition of unsupervised learning of the autoencoder.
3) Word2vec layer:
the learner preference discipline focuses on content for the learner on the platform. However, simply comparing the number of similarities or differences between the learner's preference discipline vectors obviously does not meet the similarity factor requirement. The degree of similarity between two learners who prefer physics and prefer mathematics, but differ in terms of the subjects, is somewhat greater than two learners who prefer physics and prefer literature. The degree of association is determined by the nature of the preference discipline itself. Therefore, the embodiment of the invention determines to adopt a word2vec mode to calculate the preference subject distance between learners as similarity factor investigation.
In step S5, the current partners of the learner in the converged state in the cooperative learning social relationship neural network are basically ranked based on the scale-free network theory and the social triangle theory, and the current partner recommendation list is supplemented according to the long-tail theory of the recommendation system.
Further, in an embodiment of the present invention, step S5 specifically includes:
step S501, examining the current social network behavior in the learner collaborative learning social relationship neural network, determining whether the basic scale-free network characteristics and the evolution mode are met, and if so, executing step S502;
step S502, the evolution of the current partner network is carried out according to the probability connection and enhancement mode of the scale-free network until the convergence state is reached;
and S503, based on the social triangle stabilization principle, basically sequencing the buddies of the current learner according to the current buddies of the learner in the convergence state, and supplementing the current buddy recommendation list according to the long-tailed theory of the recommendation system.
The embodiment of the invention also comprises the following steps:
in step S6, basic information data of the learner is tracked to periodically update learner attributes and a partner network structure, and new learner recommended partnership calculation is performed according to the new learner personalized attributes.
Specifically, as shown in fig. 6, according to the non-scale-small world basic social network theory, the embodiment of the present invention uses the more stable Jaccard Distance to determine the basic Distance between the user pairs. The closer the distance the more likely the person will interact in the following process. And simulating the next interactive process by adopting random numbers. And if the interaction is carried out, the edge weight is increased by 1, and the interaction probability of the next round is further improved according to the social aggregation coefficient. Otherwise, the edge weight is reduced according to the social relationship elimination rate. And when the edge weight reaches below 0.5, judging that the current user basically does not interact, and deleting the opposite edge of the user. Until the final edge number of the full graph is almost unchanged to reach convergence. In order to prevent the learner from learning only in a small circle, and in combination with the social trigonometry, the embodiment of the invention considers the additional users into the recommended buddy list again, and the general flow is as shown in fig. 7. After the partners of the convergence partner are inverted according to the cosine distance, two users are selected to be added into the recommendation partner list of the current learner according to the classical rephoto coefficient of 0.2 of the recommendation method in the embodiment of the invention.
Finally, the embodiment of the invention finally adopts the algorithm to combine with an online education platform to realize the learner online communication learning platform based on the system learning social relationship recommendation, and as shown in fig. 8, the system is divided into a data collection module, an algorithm execution module, a data persistence module and a basic front-end display module.
To sum up, the method for constructing a social relationship for learner collaborative learning for online education according to the embodiments of the present invention introduces a logical architecture and a technical process of feature engineering and social network analysis, and adopts a method of multidimensional quantization, graph structure convergence and neural network refined evaluation to generate basic feature attributes of learners on a behavior sequence and a speech record in a discussion area on a platform, and utilizes attribute calculation and matching to construct the social relationship for learner collaborative learning, so that the learning process of learners in the platform is not performed independently, and more efficient learning is completed through interaction between different learners, thereby pertinently solving the problems of low interaction degree between learners and difficulty in discovering good learning partners in the current online education platform, and recommending a suitable learning partner for learners by analyzing the specific behavior sequence and specific features of learners on the platform to enhance the learning effect on the platform.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
In the description of the specification, reference to the description of "one embodiment," "some embodiments," "an example," "a specific example," or "some examples" or the like means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (7)
1. A construction method for learner collaborative learning social relations facing online education is characterized by comprising the following steps:
s1, acquiring basic information data of a learner in an online education platform, and extracting attribute characteristics of the basic information data by adopting a sequence analysis method, wherein the attribute characteristics comprise classified learner behavior characteristics, basic distances among disciplines and interactive positive data;
s2, performing behavior analysis on the basic behavior characteristics in the attribute characteristics to label the partnership;
s3, processing the partnership labels by using a k-Means clustering algorithm to obtain the clustering discovery of the learning community of the learner;
s4, constructing a general neural network according to the learner learning community clustering discovery, and training the learner collaborative learning social relationship neural network by using the attribute characteristics to obtain the learner collaborative learning social relationship neural network so as to obtain a current buddy recommendation list;
and S5, basically sequencing the current partners of the learner in the cooperative learning social relationship neural network in a convergence state based on a scale-free network theory and a social triangle theory, and supplementing the current partner recommendation list according to a long-tail theory of a recommendation system.
2. The method as claimed in claim 1, wherein the step S1 includes:
step S101, obtaining initial information data of a learner in an online education platform, and performing null filling and basic unique heat treatment on the basic information data to obtain basic information data, wherein the basic information data comprises a behavior sequence, a community interaction behavior record, subject preferences, learning time, a learning place and a system login mode;
step S102, extracting the basic information data by adopting a sequence analysis method to obtain basic behavior characteristics and a corresponding vectorization result;
step S103, processing the basic behavior characteristics, the corresponding vectorization results and the community interaction behavior records according to a psychological personality dividing and scoring mechanism to obtain a personality characteristic value of the learner;
step S104, carrying out weighted average numerical calculation on the basic behavior characteristics and the learner character characteristic numerical values to obtain a primary numerical value, taking the average social activity and platform activity of the calculated users as the behavior characteristics of the learner, and carrying out Gaussian distribution and classification on the behavior characteristics of the learner according to the characteristics of Gaussian distribution of behavior attributes in the crowd to obtain classified behavior characteristics of the learner;
step S105, carrying out quantification processing on learner subject preferences in the basic information data by adopting word2vec to obtain basic distances among the subjects;
and step S106, performing word segmentation analysis on the community interaction behavior record in the basic information data by adopting a word segmentation tool, and performing numerical quantification processing on whether the current interaction is positive according to Chinese word emotion values to obtain interaction positive data.
3. The method as claimed in claim 1, wherein the step S2 includes:
step S201, dividing and filtering the basic behavior characteristics in the attribute characteristics to obtain the real interactive behaviors of the learner and the non-interactive behaviors of the learner;
step S202, analyzing the real interactive behavior of the learner and the no interactive behavior of the learner to derive the partnership label.
4. The method as claimed in claim 1, wherein the step S3 specifically includes:
step S301, defining the learner partner interactive behavior in the partnership labeling;
step S302, calculating the Jaccard Distance between learners according to the defined learner partner interactive behaviors;
and step S303, processing the Jaccard Distance between learners by using the k-Means clustering algorithm to obtain the learning community clustering discovery of learners.
5. The method as claimed in claim 1, wherein the step S4 specifically includes:
step S401, obtaining a time series relationship through learner learning community clustering discovery, dividing a learner behavior sequence according to the time series relationship, and obtaining a recommendation buddy list through calculation of a learner sequence with later time and a learner sequence with earlier time;
step S402, interactive behaviors between the learner and other users are obtained through learner learning community clustering discovery so as to label the partnership of the learner and the other users;
and S403, constructing a general neural network according to the recommendation partner list and the labeled partnership of the recommendation partner list and the labeled recommendation partner list based on deep learning, training the general neural network by using the attribute characteristics to obtain a learner collaborative learning social relation neural network, and further obtaining a current partner recommendation list.
6. The method as claimed in claim 1, wherein the step S5 includes:
step S501, examining the current social network behavior in the learner co-learning social relationship neural network, determining whether the basic scale-free network characteristics and the evolution mode are met, and if so, executing step S502;
step S502, evolution of the current partner network is carried out according to the probability connection and enhancement mode of the scale-free network until the convergence state is reached;
step S503, based on the social triangle stabilization principle, the method basically orders the buddies of the current learner according to the current buddies of the learner in the convergence state, and supplements the current buddy recommendation list according to the long-tailed theory of the recommendation system.
7. The construction method of online education-oriented learner co-learning social relationships according to claim 1, further comprising:
and S6, tracking basic information data of the learner to periodically update the characteristic attribute and the partner network structure of the learner, and performing new learner recommended partner relation calculation according to the new personalized characteristic attribute of the learner.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110730170.5A CN113378076B (en) | 2021-06-29 | 2021-06-29 | Online education-oriented learner collaborative learning social relationship construction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110730170.5A CN113378076B (en) | 2021-06-29 | 2021-06-29 | Online education-oriented learner collaborative learning social relationship construction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113378076A CN113378076A (en) | 2021-09-10 |
| CN113378076B true CN113378076B (en) | 2022-10-18 |
Family
ID=77579966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110730170.5A Active CN113378076B (en) | 2021-06-29 | 2021-06-29 | Online education-oriented learner collaborative learning social relationship construction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113378076B (en) |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8494978B2 (en) * | 2007-11-02 | 2013-07-23 | Ebay Inc. | Inferring user preferences from an internet based social interactive construct |
| US8666909B2 (en) * | 2007-11-02 | 2014-03-04 | Ebay, Inc. | Interestingness recommendations in a computing advice facility |
| US20140188575A1 (en) * | 2012-12-31 | 2014-07-03 | Laureate Education, Inc. | Collaborative quality assurance system and method |
| US9324038B2 (en) * | 2013-11-15 | 2016-04-26 | Xerox Corporation | Method and system for clustering, modeling, and visualizing process models from noisy logs |
| CN105404687A (en) * | 2015-12-11 | 2016-03-16 | 重庆市科学技术研究院 | Personalized recommendation method and system for learning behavior |
| CN108182489B (en) * | 2017-12-25 | 2021-06-18 | 浙江工业大学 | Personalized learning recommendation method based on online learning behavior analysis |
| CN108537636A (en) * | 2018-04-08 | 2018-09-14 | 林凡 | Online collaborative filtering commending contents algorithm based on user's dynamic behaviour similitude |
| CN108629630B (en) * | 2018-05-08 | 2020-05-12 | 广州太平洋电脑信息咨询有限公司 | Advertisement recommendation method based on feature cross-combination deep neural network |
| US20190384255A1 (en) * | 2018-06-19 | 2019-12-19 | Honeywell International Inc. | Autonomous predictive real-time monitoring of faults in process and equipment |
| CN109636495A (en) * | 2018-09-21 | 2019-04-16 | 闽南理工学院 | A kind of online recommended method of scientific and technological information based on big data |
| CN109582875B (en) * | 2018-12-17 | 2021-02-02 | 武汉泰乐奇信息科技有限公司 | Personalized recommendation method and system for online medical education resources |
| CN109816015B (en) * | 2019-01-22 | 2021-01-08 | 北京科技大学 | Recommendation method and system based on material data |
| CN110968426B (en) * | 2019-11-29 | 2022-02-22 | 西安交通大学 | Edge cloud collaborative k-means clustering model optimization method based on online learning |
| CN111460249B (en) * | 2020-02-24 | 2022-09-09 | 桂林电子科技大学 | Personalized learning resource recommendation method based on learner preference modeling |
| CN111401525A (en) * | 2020-03-20 | 2020-07-10 | 珠海读书郎网络教育有限公司 | Adaptive learning system and method based on deep learning |
-
2021
- 2021-06-29 CN CN202110730170.5A patent/CN113378076B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113378076A (en) | 2021-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rasheed et al. | Learning style detection in E-learning systems using machine learning techniques | |
| Bauman et al. | Recommending remedial learning materials to students by filling their knowledge gaps | |
| Chen et al. | Mobile formative assessment tool based on data mining techniques for supporting web-based learning | |
| Huyge et al. | Does students' machismo fit in school? Clarifying the implications of traditional gender role ideology for school belonging | |
| CN109902912B (en) | Personalized image aesthetic evaluation method based on character features | |
| Van Rossem | Introducing a cognitive approach in research about generational differences: The case of motivation | |
| CN113656687B (en) | Teacher portrait construction method based on teaching and research data | |
| Capuano et al. | A personality based adaptive approach for information systems | |
| Wei et al. | Personalized online learning resource recommendation based on artificial intelligence and educational psychology | |
| Cam et al. | Discovery of Course Success Using Unsupervised Machine Learning Algorithms. | |
| CN113378076B (en) | Online education-oriented learner collaborative learning social relationship construction method | |
| Bouzayane et al. | Prediction method based DRSA to improve the individual knowledge appropriation in a collaborative learning environment: case of MOOCs | |
| Ozturk et al. | Relationship between learner profiles and learner–content interaction in online learning: Exploring implications for learning experience design | |
| Vahdat | Learning analytics and educational data mining for inquiry-based learning | |
| Ma et al. | Design a course recommendation system based on association rule for hybrid learning environments | |
| Zhang et al. | Construction and research of constructive English teaching model applying multimodal neural network algorithm | |
| PEROVIĆ | Data mining influence on e-learning | |
| Sayed et al. | Predict student learning styles and suitable assessment methods using click stream | |
| Sun et al. | Study on the distance learners’ academic emotions using online learning behavior data | |
| KR102599368B1 (en) | System for customized content providing based on big data and method the same | |
| Hui | Continual Construction of Adaptive Learning Model for English Vocabulary Using Machine Learning and Virtual Reality | |
| Rizzello | Identity and consciousness in economics | |
| Shanthi et al. | Recommender System for Predicting Students' Academic performance in association with Cognitive state and Affective state using Sentiment Analysis and Association Rule Mining on the closed ended questionnaire | |
| Gawande et al. | A survey to enhance learning management systems to increase productivity of the students and their performance | |
| Peng et al. | Personalized Learning Resource Recommendation Based On Learner Profile |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |