International Journal of Information Technology (IJIT) – Volume 6 Issue 4, Jul-Aug 2020 ISSN: 2454-5414 www.ijitjournal.org Page 44 Identification of Sentiments and Opinions of Twitter Data Nikita R. Dandwate [1] , Sarika B. Solanke [2] Department of Computer Science & Engg. Deogiri Institute of Engg. & Management Studies, Aurangabad. ABSTRACT In this paper we proposed a model for Sentiment Analysis and Opinion Mining on the collected Twitter Data. Twitter sentiment analysis has recently emerged as a hot research topic in last few years. Most of existing solutions to Twitter sentiment analysis only consider textual information of Twitter messages, and probably fails when short or confusing messages or conversations appear. Sentiment analysis is defined as the category of natural language processing-based and computational technique. It is then used to sense, extract and illustrate information, which is subjective, and is expressed in a given part of text. The main intention of sentiment analysis is to classify the writer’s attitude towards different topics into various categories like positive, negative or neutral. Current studies display that sentiment diffusion patterns on Twitter have very close relations with sentiment polarities of the Twitter messages. Hence, in this research paper we focus on how to fuse textual information of Twitter messages and sentiment Analysis patterns to obtain better performance on Twitter data. To the end of this paper we will have a properly defined system that will help us to classify the collected tweets based on the sentiments associated to it and thus help in developing a reliable system that can be used by various teams for their analysis purpose to understand the sentiments of people towards some trending facts and help them in further decision making process. Keywords: - Twitter Data, Sentiment Analysis, Aspects, Sentiment Polarities, Opinion Mining, Social Network, Text Classification, Gibbs Sampling Algorithm, N-Gram Algorithm, LDA Topic Modelling. I. INTRODUCTION Our world has been thoroughly transformed by dig- ital technology. A research shows that nearly all Internet users go online to conduct some of their ordinary day-to-day activities, from mundane tasks to social arrangements to personal recreation. The data society is characterized by ever growing volumes of material. Driven by the current generation of Web applications, the nearly limitless connec- tivity, and an insatiable desire for sharing information, par- ticularly among younger generations, the volume of user- generated social media content is growing rapidly and likely to increase even more in the upcoming years [2]. Textual information in the world can be broadly categorized into two main types: facts and opinions. Facts are objective ex- pressions about entities, events, and their properties. Opin- ions are usually subjective expressions that describe peo- ple’s sentiments, appraisals, or feelings toward entities, events, and their properties. The concept of opinion is very broad. Opinions are so important that whenever we need to decide, we want to hear other’s opinions. Opinions are cen- tral to almost all human activities and are key influencers of our behaviors. Our beliefs and perceptions of reality, and the choices we make, are, to a considerable degree, conditioned upon how others see and evaluate the world. For this reason, when we need to decide, we often seek out the opinions of others. “What other people think” has always been an im- portant piece of information for most of us during the deci- sion-making process [1]. Sentiment analysis systems have found their applications in almost every business and social domain. One such application of Sentiment Analysis applies to Twitter Data. Twitter, a popular micro-blogging service around the world, has been shaping and transforming the way people obtain information from people or organizations that they are interested in. Since established in 2006, Twitter has become one of the largest online social networking plat- forms in the world. Given the ever-growing amount of data available from Twitter, mining users’ sentiment polarities expressed in Twitter messages has become a hot research topic due to its wide applications. For example, by analyz- ing Twitter users’ sentiment polarities on political parties and candidates, several tools have been developed to pro- vide strategies for political elections. Business companies also use Twitter sentiment analysis as a fast and effective way to monitor people’s feelings towards their products and brands. The objective of sentiment analysis on Twitter data is to classify the sentiment polarity of a Twitter message as positive, neutral or negative. There are more slangs, acro- nyms, misspelled words and modal particles in Twitter mes- sages due to their casual form. As a result, the performance of traditional text sentiment analysis algorithms drops dras- tically when applied to predict sentiment polarities of Twit- ter messages. To solve this problem, many novel sentiment analysis methods for Twitter messages have been developed. These methods can be roughly divided into two categories: fully supervised methods and distantly supervised methods. The fully supervised methods aim to learn sentiment classi- fiers based on manually labelled data and sentiment lexi- cons. One major problem of fully supervised methods is that it is time-consuming and labor-intensive to manually build sentiment lexicons and label the data, and consequently the RESEARCH ARTICLE OPEN ACCESS