Abstract
Sentiment Classification refers to the computational techniques for classifying whether the sentiments of text are positive or negative. Statistical Techniques based on Term Presence and Term Frequency, using Support Vector Machine are popularly used for Sentiment Classification. This paper presents an approach for classifying a term as positive or negative based on its average frequency in positively tagged documents in comparison with negatively tagged documents. Our approach is based on term weighting techniques that are used for information retrieval and sentiment classification. It differs significantly from these traditional methods due to our model of logarithmic differential average term distribution for sentiment classification. Terms with nearly equal distribution in positively tagged documents and negatively tagged documents were classified as a Senti-stop-word and discarded. The proportional distribution of a term to be classified as Senti-stop-word was determined experimentally. Our model was evaluated by comparing it with state of art techniques for sentiment classification using the movie review dataset.
Highlights
The web which is massively increasing resource of information has changed from read only to read write
Our approach is based on traditional techniques of Information Retrieval, we examine whether addressing sentiment classification as special case of information retrieval can improve classification accuracy
From the results of the experiments conducted it can be observed that accuracy of AVERAGE RELATIVE TERM FREQUENCY SENTIMENT CLASSIFIER (ARTFSC) is more than Delta TFIDF
Summary
The web which is massively increasing resource of information has changed from read only to read write. Organizations provide opportunity to the user to express their views on the products, decisions and news that are released [1]. Users can express their emotions as well can comment on the earlier user sentiments. Large amount of sentiment data is generated by various users for different features of products and services. Processing this sentiment data needs to be handled systematically
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
