Social media analysis and summarization for opinion mining: a business case study

A huge amount of social media data is generated daily as a result of the interactions between subscribers of the social networking platforms. Subscribers are characterized by diversity in background, geographical locations, opinions, etc. Further, the diversity and pervasiveness of the data together with its social characteristics make it an intriguing source for gauging the public opinion. Twitter is one of the most famous and popular networking platforms. It provides a micro-blogging environment in which subscribers can post and read short messages called Tweets. In this paper, we propose a method for finding groups of people who express similar ideas about a specific subject in their postings. We also introduce an algorithm to label each group of users based on their postings. The labels help us to get a more profound insight into the main opinion of each group. The work described in this paper has high significance and a variety of practical applications. Applicability and effectiveness of the proposed approach have been demonstrated in a case study on the Tweets related to Royal Bank of Canada. The reported results prove the viability and efficiency of our approach. Besides, the proposed approach is generic enough to be adapted and applied to any type of existing social media platforms to extract main ideas related to a specific subject.