Using Textual Similarity and Sentiment Analysis in Discussions Forums to Enhance Learning

다국어 초록 (Multilingual Abstract)

Web 2.0 technologies have changed the learning process and have affected the unidirectional relationship obtaining between the teacher and the student. Web 2.0 has given rise to new bidirectional learning strategies in which the learner is both an actor and active member in the learning process. Given the wide use of these technologies as educational tools, our objective in this article is to analyze learners’ comments in discussion forums, which are popular platforms amongst students. For this purpose, we use the Latent Semantic Analysis (LSA), a statistical model that can allow us to establish the relationship between the set of messages published by learners and course content that students need to master. We also use sentiment analysis techniques to identify the students’ opinions about educational issues that are problematic. The implementation of both methods has revealed that the data available in educational discussion boards can improve the quality of learning and teaching.

목차 (Table of Contents)

• Abstract
• 1. Introduction
• 2. Discussion Forums and e-Learning
• 3. Towards a Discussion Forum that Supports Learning
• 3.1. Preprocessing
• 3.2. Keywords Extraction
• 3.3. Representation
• 3.4. Textual Similarity
• 3.5. Sentiment Analysis
• 3.6. Algorithm
• 3.7. Evaluation
• 4. Simulation and Results
• 4.1. Corpus Description
• 4.2. Results and Discussion
• 5. Conclusion
• References