Sentiment Analysis of Mobile Phones Using Revised Data Dictionary and Polarity

Abstract

Twentieth century s one of the most powerful discovery is social media. Simply, it is a platform where numerous people can communicate with each other, can share their thoughts or opinion. Though initially, its use was just communication, later on, significant newlineenhancements are seen, such as advertisement, marketing, video entertainment, business development etc. Opinion mining is defined as the function of classifying texts into newlinecategories, depending on whether they express positive or negative emotions, or do they give no sense. The power of social media to reflect and influence public opinion and influence of social media has made them an area of great interest for marketers, communications experts and companies who want to advertise their products and services, or just their brand want to promote and monitor the name. They will always look for a product that is having good reviews and avoid that product that having the bad reviews. newlineThe work will be focused on improving data dictionary by inserting (positive dictionary, negative dictionary, slangs dictionary, abbreviations dictionary) with respect to English language to enhance data preprocessing task so that we will have more accurate results. Further for more accurate result, we have used, the basic concept of polarity of a data for newlinerecognizing whether sentence is positive, negative or neutral. The main aim is to find out the polarity or idea, that the statement has something positive or very positive. Polarity is the way through which we can get the accurate reviews about the product. It not only classify the words into positive, negative or neutral but it provides the range by which newlinewords can be categorized into strong positive or strong negative Now, we are going to discuss on Sentiment Index and Sentiment Score. Sentiment Index is newlinethe ratio of difference of positive emotions and negative emotions to the sum of positive emotions, negative emotions and neutral emotions of users or organizations and marketers to understand the day to day user requirements.