Sparse learning strategy and key feature selection in interactive learning environment

ABSTRACT

Effective analysis and demonstration of these data features is of great significance for the optimization of interactive learning environment and learning behavior. Therefore, we take the big data set of learning behavior generated by an online interactive learning environment as the research object, define the features of learning behavior, and demonstrate their relationships. A key feature selection method for sparse learning based on data set is designed. The models and algorithms are fully trained and tested through a large number of experiments. Several approximate optimal algorithms are selected to compare the performance indicators. On this basis, the rule and relationships are mined and predicted for the key features, and the measures to improve the key features are proposed. The conclusion is that the guidance and construction of learning behavior based on the key features can have a significance on the learning effect, that has also been proved in practice. Driven by the actual data, it is an inevitable trend to design the suitable methods applied to key features of education big data. This research method and practice process can provide technical reference and theoretical basis for the similar topics.

KEYWORDS:

• Interactive learning environment
• sparse learning
• feature selection
• learning analytics
• education big data

Acknowledgements

Thanks for the technical support provided by the laboratory of School of Software of Tsinghua University, as well as the theoretical guidance and practical reference provided by Qufu Normal University.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study is supported by Shandong Education Science Plan (grant No. 2020ZD030) and National Office for Philosophy and Social Sciences (grant No. BEA190107).

Notes on contributors

Xiaona Xia

Xiaona Xia is an associate professor and PhD of Qufu Normal University. She is also a member of IEEE computer society and CCF. Her research interests include learning analytics, collaborative Learning, education big data, educational statistics, computer supported learning, data mining, service computing, etc.