Facebook as a Corpus for Emoticons-Based Sentiment Analysis
Geetika Vashisht, Sangharsh Thakur · 2014
Abstract — With the growing popularity of the social networking sites, usage of informal language, short cuts & emoticons is increasing rapidly. The use of emoticons in text in order to express sentiments is posing a challenge to the automated sentiment analysis tools to correctly account for such graphical cues for sentiment. This paper aims at demonstrating how emoticons typically convey sentiments and how we can exploit emoticons by using a manually created emoticon sentiment lexicon and then using finite state machines to find out the polarity of the sentence or paragraph. We evaluate our approach on 1,250 Facebook status and 2,050 Facebook comments, which all contain emoticons and have been manually annotated for sentiment. We identified the most commonly and frequently used emoticons & classified them on the basis of the sentiment they strengthen which eventually decides the polarity of the sentence. In this paper we want to introduce a method to perform a sentiment analysis on text-based status updates & comments, disregarding all verbal information and using only emoticons to detect both positive and negative sentiments.