Multi-aspect and Multi-class Based Document Sentiment Analysis of Educational Data Catering Accreditation Process

N.D. Valakunde,M.S. Patwardhan

DOI: https://doi.org/10.1109/cube.2013.42

2013-11-01

Abstract:Sentiment analysis is used to determine the attitude of a writer with respect to some topic or the overall contextual polarity of a document. Current techniques of sentiment analysis work either at a document or at a sentence or at an entity level. The objective of this research is to compute the document level sentiment analysis score based upon underlying entity or aspect based scores, and their importance towards the evaluation of the topic discussed in the complete document. In this approach we have classified the document into multiple classes with multiple aspects are taken into consideration. The approach allows us to automatically take care of the current problems of document level sentiment analysis, such as, entity identification, subjectivity detection and negation. The technique is further applied for educational data mining, where a faculty performance is evaluated using the sentiment analysis of comments provided by students as a part of their feedback. The document level faculty performance score is computed from the distinct aspect based sentiment scores, such as, knowledge, presentation, communication and regularity of the faculty. The importance of these aspects towards computation of document level score is taken based upon the weightages. These aspects have, taken up from the NAAC accreditation criteria. The results show that this strategy provides more accurate document level sentiment scores than the scores computed directly at the document level without analyzing the underlying aspects.