期刊名称:Journal of Theoretical and Applied Information Technology
印刷版ISSN:1992-8645
电子版ISSN:1817-3195
出版年度:2014
卷号:69
期号:1
出版社:Journal of Theoretical and Applied
摘要:E-learning is becoming the most influential and well-liked standard for learning through web based education. It is very important to categorize the online feedback of the learners emotion in e-learning system. Learning usually refers to teaching skills propagated with the help of computers to communicate knowledge in a web based classroom environment. It is very difficult to identify the learner�s emotional state whether they are satisfied with the online courses. The twitter sentiment mining framework helps to find about the learners who are frequently interacting with the e-learning environment. Twitter has become the most popular micro-blogging area recently. Millions of users frequently share their opinion on the blogs. Twitter is referred as a right source of information to perform sentiment mining. This research presents a new method for sentiment mining in twitter based messages written by learners, initially helps to extract information about learners sentiment polarity (negative, positive), and to model the learners sentiment polarity to identify the change in their emotions. A model has been constructed from the training data of the sentimental behaviors of the e-learners using Na�ve Bayesian approach. The model constructed has been tested through the test data during the prediction process of discovering the emotional states of e-learner. The results were compared against the other famous classification algorithms like support vector machines and maxentropy techniques. This information can be effectively used by e-learning system, by considering the learners' emotional state when recommending learner�s the most appropriate activity each time. The learner�s sentiment, emotional state towards the online course can provide feedback for e-learning systems. The experimental outcome show that our proposed research work outperforms recent supervised machine learning algorithms on accuracy findings of learner�s emotional state classification.