International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056 Volume: 03 Issue: 08 | Aug-2016 www.irjet.net p-ISSN: 2395-0072 © 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 1332 A Study on Various Classification Techniques for Sentiment Analysis on Social Networks N.SARANYA MSc (SS), MCA, M.Phil, 1 , Dr.R.GUNAVATHI 2 1 Assistant Professor, PG Department of Computer Science, Sree Saraswathi Thyagaraja College, TN, India. 2 Head,Department of Computer Application, Sree Saraswathi Thyagaraja College, TN, India. -------------------------------------------------------------------***---------------------------------------------------------------------- Abstract - Sentiment analysis, which is additionally called as opinion mining, involves in building a system to gather and examine opinions regarding the merchandise created in journal posts, comments, reviews or tweets. Sentiment Analysis is the study of automatic identification of online user’s preference and classification of these preferences into its positive, negative or neutral orientation. A major problem occurs when we try to determine the sentiment or the class of these data i.e. whether the data is good or bad. Analyzing the sentiment of a text, document or an article is a challenging task in the world. Several methods were implemented for sentiment analysis throughout the years, but still more improvement and perfection is needed. Here in this paper various sentiment analysis techniques are reviewed. Key Words: Opinion Mining, Classification Techniques, Sentiment Analysis, Sentiment Classification, Summarization. 1. INTRODUCTION E-commerce and the rapid growth of the social media, individuals and organizations are progressively using the content on these media for decision making purpose [1], [2]. Hence there is a need for effective and accurate automated approach which predicts the opinions presented in the reviews. Opinion mining systems classify opinion data available on web into their respective opinion polarity of positive, negative or neutral. In the current web era, opinion mining is one of the widely studied topics under Web Mining and Natural Language Processing. Opinion mining is analyzed at the document level, sentence level and aspect level. This survey aims to give a closer look on these enhancements and to summarize and categorize some articles presented in this field according to the various Sentiment Analysis (SA) and Sentiment Classification (SC) techniques. The sentiment classification techniques, as shown in Figure 1 are discussed with more details illustrating related articles and originating references as well. Figure 1 - Sentiment classification techniques. 2. RELATED WORKS Opinion mining is a popular research topic because of its wide range of application as well as various challenging research problems involved. It is studied under various fields including Data Mining, Information Retrieval, Web Mining and Natural Language Processing. We now present some of the related work carried out. Mining Hu and Bing Liu [3] aim to summarize all the client reviews of a product. This report task is totally different from ancient text report as a result of the others area unit solely curious about the precise options of the merchandise that customers have opinions on and conjointly whether or not the opinions area unit positive or negative. The others don't summarize the reviews by choosing or revising a set of the initial sentences from the reviews to capture their small print as within the classic text report. During this paper, the others solely specialize in mining opinion/product options that the review the others have commented on. variety of techniques area unit given to mine such options. They proposed a technique where