| Summary: | 碩士 === 國立臺南大學 === 數位學習科技學系碩博士班 === 107 === Based on the collaborative learning process of the past research, this study developed an affective Chat-Bot system based on the Retrieval-based Model and Neural Network. The emotional classification is based on Levenshtein Distance and The Long Short-Term Memory algorithm. The research proposes a specific framework and analysis method for letting affective Chat-Bots and college students participate in computer-supported collaborative learning, in order to explore the learning style and pattern. In the first part of the study, learning effectiveness was investigated and quantified using knowledge assessment tests and questionnaires, and analyzed and compared by data statistics. In the second part of the study, using the Awareness Tools and Text Mining's TF-IDF method, this study analyzes the behavioral patterns, emotional states, and mental processes of students' online collaboration. The results show that the Chat-Bot significantly increases the heat of group discussion and the centripetal force between members, making the subjects more active in reflection, criticism, problem solving and creativity. However, there was no significant difference between the experimental group and the control group in the knowledge test scores. In addition, the interview analysis shows that the participation of robots will have an important impact on students' learning emotions and self-awareness. As a reference and guidance for follow-up research, human-computer interaction design based on artificial intelligence is essential before the experiment.
|
|---|
