Chapter 3 - Social media data analytics using feature engineering

Abstract

In this fast-growing digital world, social media analytics is gaining attention in the field of big data. Big data is the collection of huge amounts of raw digital data that is difficult to analyze with conventional analysis methods. Due to the popularity of social media sites such as Twitter and Facebook, a vast amount of public unstructured data is generated by millions of users every day. This raw bulk data cannot be directly used in decision making and prediction tasks. Therefore many researchers have been working on converting this huge unstructured information into meaningful information through big data analytics. Developing an efficient data analytics tool is essential to understand the multimodal social media data that improves the performance of decision-making and prediction systems. This chapter overviews the strengths of the current state-of-the-art feature engineering approaches that have been developed to represent the characteristics of social media data. In addition, this chapter outlines the proposed framework for analyzing the social media data including the shared images, tags, associated comments, and its social relationship.