Modeling and Analysis of Learners’ Emotions and Behaviors Based on Online Forum Texts
Mingyong Li, Mingyuan Ge, Honggang Zhao, Ziye An
Abstract
Under the continuous impact of the epidemic, online learning methods represented by MOOC have developed rapidly. The course forum area has produced a large amount of text-based unstructured data, which can reflect the potential characteristics of learners' emotional states and behavioral interactions, and has an important impact on students' learning outcomes. To this end, this paper constructs an emotional and behavioral analysis model based on online forum texts, obtains forum data from the "Python Language Programming" course on the Chinese University MOOC platform, uses domain dictionary emotion classification method to analyze learning emotions, and based on the method of cognitive behavior coding table and knowledge construction behavior coding table analyzes learners' cognitive behavior and knowledge construction behavior. It can dynamically analyze learners' emotions, behavior changes, and evolutionary trends. This research provides opinions and suggestions on the improvement of platform interactive functions for teachers' online teaching, students' online learning, and platform management, which can effectively improve the efficiency and effectiveness of online learning.