A SURVEY ON CONCEPTUAL ADVANCEMENTS IN SENTIMENT ANALYSIS

Devesh Rathi B, Shinde Shrutika S, Dhoka Shubham R, Jagruti Mahajan R

Abstract: Sentiment analysis is defined as the process of computationally identifying and categorizing the opinions expressed in the written text to analyze the writer’s attitude and finding out whether the written text is positive, negative or neutral in nature. Over the past few years, it attracted a great deal of attentions from both academia and industry due to many challenging research problems and a wide range of applications. These opinions are generally obtained in text format through various databases such as twitter, customer forum, and complaint websites. In these survey paper we have studied and shown the advancements of the sentiment analysis overs the past few years and have inhabited the growth of various concepts under it. This paper gives a review of various facts about the different perceptions related to sentiment analysis and opinion mining. We have also suggested a system as a future scope which works on the analysis of various opinions and creates a module that can handle multiple entities and queries related to it.

Keywords: Sentiment analysis, Opinion mining, Hadoop Framework, Node architecture

DOI: https://doi.org/10.15623/ijret.2016.0501047