Learning Sentiment from Students’ Feedback for Real-Time Interventions in Classrooms

Abstract

Abstract. Knowledge about users sentiments can be used for a variety of adaptation purposes. In the case of teaching, knowledge about students sentiments can be used to address problems like confusion and boredom which affect students engagement. For this purpose, we looked at several methods that could be used for learning sentiment from students feedback. Thus, Naive Bayes, Complement Naive Bayes (CNB), Maximum Entropy and Support Vector Machine (SVM) were trained using real students' feedback. Two classifiers stand out as better at learning sentiment, with SVM resulting in the highest accuracy at 94%, followed by CNB at 84%. We also experimented with the use of the neutral class and the results indicated that, generally, classifiers perform better when the neutral class is excluded.

Text mining in education

“…Analytics/student information This paper utilized a method of opinion mining to help e-learning systems to know the users' opinions on the course-wares. (Altrabsheh, Cocea, & Fallahkhair, 2014) User feedback…”

Un modelo basado en el Clasificador Naïve Bayes para la evaluación del desempeño docente

“…Asimismo, hay investigaciones (Altrabsheh, 2014;Ortigosa, 2014) para monitorear y analizar en tiempo real los comentarios que los estudiantes hacen en redes sociales con el fin de mejorar las clases de los docentes. Además, algunos autores (Bravo, 2013) documentan el uso de la red social Twitter como una herramienta complementaria en la educación colaborativa y el aprendizaje informal en el área de ingeniería.…”

Sentiment discovery and analysis as a mean of student experience improvement

“…Sentiment analysis has found its application in certain service sectors, such as financial, health or tourism, but only few research has been conducted on the application of sentiment analysis in higher education, such as Hosterman 2013, (Lewis & Nichols, 2013), Yaros (2013), (Zeng, Hall, & Jackson Pitts, 2013), (Altrabsheh, Cocea, & Fallahkhair, 2014), (Altrabsheh, Gaber, & Cocea, 2013), (Marques, Krejci, Siqueira, Pimentel, & Braz, 2013). The majority of research efforts in the area of higher education aimed at improving intelligent tutoring systems.…”

“…In the paper (Altrabsheh, Cocea, & Fallahkhair, 2014), the authors examined how knowledge about the sentiment of students who are at risk of quitting a course primarily due to boredom or confusion, may be used to encourage their involvement in a teaching process. It was demonstrated on the collected data set, that elimination of the neutral sentiment class in the training phase produced better results, and that the Support Vector Machine and the Complement Naive Bayes classifiers were the best analytic methods for a given data set.…”