Author(s): Xin Wan | Qimanguli Jamaliding | Toshio Okamoto
Journal: Journal of Computers
ISSN 1796-203X
Volume: 6;
Issue: 2;
Start page: 254;
Date: 2011;
Original page
Keywords: social interaction | Markov chain model | recommender system | group learning
ABSTRACT
Recommender systems are now a popular research area and have become powerful tools to present personalized offers to users in many domains (e.g. e-commerce, e-learning). In this paper, we introduced an approach of personalization which extracts learners’ relationship based on learning processes and learning activities (e.g. note taking) to provide more authenticity, personalized recommendations for group learning support.Base on learners’ learning activities some interaction factors are extracted by using natural language process technologies and data mining automatically. Then, extracted interaction factors are utilized to generate some relationship indicators for inferring the learners’ directive relationship. These indicators are as symbols in order to describe a situation and relative degree which knowledge and understanding are socially distributed among group learners. Thirdly, we use a machine learning approach for acquiring a learner relationship identify module according to the relationship indicators.The experimental result shows that the proposed approach can give a more satisfying and qualified recommendation.
Journal: Journal of Computers
ISSN 1796-203X
Volume: 6;
Issue: 2;
Start page: 254;
Date: 2011;
Original page
Keywords: social interaction | Markov chain model | recommender system | group learning
ABSTRACT
Recommender systems are now a popular research area and have become powerful tools to present personalized offers to users in many domains (e.g. e-commerce, e-learning). In this paper, we introduced an approach of personalization which extracts learners’ relationship based on learning processes and learning activities (e.g. note taking) to provide more authenticity, personalized recommendations for group learning support.Base on learners’ learning activities some interaction factors are extracted by using natural language process technologies and data mining automatically. Then, extracted interaction factors are utilized to generate some relationship indicators for inferring the learners’ directive relationship. These indicators are as symbols in order to describe a situation and relative degree which knowledge and understanding are socially distributed among group learners. Thirdly, we use a machine learning approach for acquiring a learner relationship identify module according to the relationship indicators.The experimental result shows that the proposed approach can give a more satisfying and qualified recommendation.
