Predicting Learner's Confidence from their Behaviour using a Judgment Questionnaire

摘要

Affective computing gives the study and development of the system that can recognize emotions of the learner. In e-learning system affecting computing plays an important role to recognize the emotional state of the learner. This paper represents how the confidence level which is an academic emotion of the learner may be predicted based on learner's behavior using judgement questionnaire. Here the confidence level rated by the learner was related to their actions in the test and tried to predict the learner's confidence in the test from their behavior. In this research all the parameters are recorded as learner's behavior during the test and the machine learning algorithms are used to classify and predict the learner's confidence. Hereby the system on its own can acts as a conventional teacher, which predicts the confidence level of the Learner based on his/her action in the online test and teaches topic accordingly. XGBoost algorithm is used in this work to train the model and to predict the confidence.