Estimating market trends by clustering social media reviews

In recent years, social media networks have expanded in size and utility. Sites like Facebook and Twitter have become giant pools of available public information. Across the globe, millions of users exchange information and sentiments daily. Realizing the importance of cyber reach, all major companies and brands have ensured their presence on all social media platforms. Companies like YouTube, Google, P&G, McDonald's continuously update their consumers pertaining to changes in their products. In this paper we have drawn an interesting relationship between the brand updates and the response garnered by them. Using the response to those updates, we have developed a causal link between the producers and consumers. We have clustered the response data on geographic parameters to see the distribution of users. To add more perspective to the information gathered, we have used sentiment analysis to distinguish positively or negatively distributed clustered data. Such a distribution gives an overview of consumer market diversity and can help the producers make better estimates of user reviews. The major contribution of this paper is introducing novel use-case of social media platforms to determine the market trends that can help investors in analyzing their performance.