Open AccessPrediction of learning behavior based on improved random forest algorithm
Author(s) -
Xiaona Xia
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1656/1/012003
Subject(s) - computer science , machine learning , artificial intelligence , random forest , predictability , set (abstract data type) , process (computing) , key (lock) , data mining , algorithm , mathematics , statistics , computer security , programming language , operating system
The research of learning behaviors is the premise of improving the learning process and results. The scale, composition and quality of the corresponding dataset will have the key impacts on research methods. Based on big dataset of learning behaviors, this study demonstrates the rules of learning behaviors, uses intelligent learning algorithms to analyze data and predict trends. Based on data characteristics, this paper evolves random forest algorithm and constructs the data structure. Taking the assessment results of learners as the test targets, the data samples are divided into two parts: train set and test set. Through the train set, we design the data model. With the help of the test set, the optimal model is finally selected, that is, the learning interaction activities are predicted and estimated. The research shows that learning behaviors have predictability and guidance, transferability and adjustability, groupness and personality. Based on this conclusion, the data analysis scheme of learning platforms is designed, which will help to improve the platform construction and learning behaviors.