A Hybrid Machine Learning Approach For Intent Classification of Domain Specific Twitter Content

Abstract

In modern society, social media has become a part of our everyday routines. Not only has it made connecting with people and building and maintaining relationships with people even in far off parts of the world effortlessly easy, it has become a platform for expressing one’s sentiments and opinions within a matter of seconds. Moreover, from a business standpoint, social media provides business organizations the opportunity to enhance brand awareness, do market research and most importantly, to establish relationships with potential as well as current consumers and improve customer relations. Social media monitoring and analyzing of social media content enables commercial brands to monitor and evaluate the reaction of the consumers to marketing campaigns and product launches in real time, to evaluate strategies and to provide exceptional customer service by responding to consumer queries and complaints effectively and efficiently. This also facilitates the policy makers and higher management of the organization a customer centric approach to monitor and evaluate the service rendered by their brand representatives regularly. Given the above reasons, analysis of social media content plays a crucial role in improving customer relations, in a way which would benefit both consumers and brand representatives. Hence, this study is an attempt at a scalable hybrid machine learning approach for intent classification of Twitter content.