Product Sales Prediction Based on Sentiment Analysis Using Twitter Data Dipak Gaikar 1 ,Bijith Marakarkandy 2 1,2 Information Technology, Thakur College of Engineering & Technology, Mumbai-400101, India Abstract—Online social media websites represent how fundamental information is created, transferred and consumed. Social media, user generated content in the form of comments, blog posts and tweets establish a connection between the producers and the consumers of information. Today’s world is connected to each other via social network like Twitter millions of people connected to each other through that network. Tracking the pulse of the social media contain, enables companies to gain feedback and insight in how to improve and market products better. It continues to offer new opportunities for organizations to directly interact with their customers or audience, the aimed at monitoring the online reputation of an organization, brand or person, social media and search engine result. This research paper uses a survey approach for movie sales prediction. This paper analyses, impact of the positive, negative, strongly positive and strongly negative online reviews of movies on the audience. It should be noted that the user feedback is given prior to watching the movie only on the basis of the online reviews. The result of this research will help the film industry to effectively address and meet the expectations of customers and stakeholder. This paper also investigates techniques for twitter data extraction using an API key. Keywords—Twitter, Sentiment Analysis, ANOVA, Prediction, Social Media, Data Mining. I. INTRODUCTION Due to its high popularity, Weblogs provide a wealth of information that can be very helpful in assessing the general public’s sentiments and opinions. It is therefore imperative to analyze them and distil useful knowledge that could be of economic values to vendors and other interested parties. Whereas marketing plays an important role for the newly released products, public opinion about the products might be crucial to determine their success in the long run. Analyzing the large volume of online reviews available would produce useful actionable knowledge that could be of economic values to vendors and other interested parties. Prediction of product performance is an extremely domain- driven task, for which a deep understanding of a variety of aspects involved are important. Previous studies have confirmed that the sentiments expressed in the online reviews are strongly correlated with the sales performance of products. From the recent studies regarding writing the reviews, online opinions, online comments, discussion forums, the most stakes is taken by film industry, which includes videos, songs, movies, television programs, etc. It is very easy to get reviews about movies after or before its release from websites dedicated for movies and it was therefore decided to take up movie as a product in this study. If the prediction is focused on electronic goods, then it is required to consider different companies/brands, but here for movies it is possible to get exact amount of the box office revenue information. There has been previous research and comparing the results with previous results was also another motivation. Various economic functions have been utilized to examine the relationship between opinions discovered from product reviews and revenue growth, stock trading volume change, as well as the bidding price variation in commercial Websites, such as eBay. Social media is increasingly being used by a large section of population in India. The content generated on social media websites have been largely untapped by businesses for gaining customer insights and predicting real outcomes. Microblogging services in recent times have been a popular communication tool among internet users. It generates millions of daily messages for popular websites. Microblogging is online word of mouth branding like Twitter, is now serving as electronic word of mouth (eWOM), forming a eWOM branding which is based on social networking and trust. Twitter has been swamped with active users during the last years and much attention has been given in analyzing the social behavior and opinions of users. The wide-spread popularity of online social networks and the resulting availability of data have enabled the investigation of new research questions, such as the analysis and estimation of public opinion on various subjects [30]. The focus of this study is on movies as it is of interest to the social media community. Large numbers of social media users discuss movies and box-office revenues this data is easily available from Internet Movie Data Base (IMDB). Microblogging sites like twitter also generate tweets which help people make decisions about watching a movie. The focus of this research is movie revenue prediction, but can be extended to other consumer products. II. RELATED WORK (Hening-Thurau etal., 2003) state that customer comments articulated via the Internet are available to a large number of other customer’s, and therefore can be expected to have a significant impact on the success of goods and services. This on consumer buying and communication behavior are tested in a large-scale empirical study. The results illustrate that consumers read online articulations mainly to save decision-making time and make better buying decisions. Structural equation modeling shows that their motives for retrieving online articulations strongly influence their behavior. Dipak Gaikar et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 6 (3) , 2015, 2303-2313 www.ijcsit.com 2303