Twitter Data Mining for Targeted Marketing

The ascendancy of social media has enabled a large number of users all around the world to share information and their beliefs regarding various topics. Various social media websites like Google+, Facebook, Instagram, Twitter, LinkedIn, etc. allow the users to share their opinions and get connected to other users in any part of the world. People openly share information about their interests in products and purchases, thus giving organizations a considerable opportunity to extract valuable insights and get an idea of customer behavior. These insights can, in turn, be used to target customers at specific locations based on their preferences. This paper aims to provide an approach to target users on Twitter by discerning their preferences with the use of unique keywords in the real estate sector. This approach consists of multiple stages that extract data from Twitter to collect usernames and carry out further classification and elimination of usernames through different stages to qualify them based on their probability of affording a property in the target regions. The accounts of qualified usernames are further crawled to find extra contact information to increase the audience reachability. This approach can thus enable organizations to send targeted advertisements to a set of users based on their locations and preferences, thus allowing them to revamp their strategy and enhance their marketing output.