A Sentiment Analysis Framework for Virtual Learning Environment

Abstract

This paper describes the development and application of a mobile framework to assist teachers in distance learning courses. The tool has a MultiAgent System (MAS) in charge of collecting and analyzing the feelings of texts posted by students in the activities forums, chats, journal and messages sent to the tutor. By means of framework it is possible to identify which students are showing negative feelings, which is the feeling that predominates in a class, to verify the history of the feelings of a certain student. As a way to evaluate the project, teachers with experience in tutoring used the framework with real course data and then to verify the efficiency of the tool, answered a questionnaire and classified postings of an VLE, as a result we obtained the accuracy of 73,88%.