Matthew Riemer

Distributed Computing in Social Media Analytics

The rise of social media has led to some of the largest cultural shifts seen so far in the twenty-first century. Billions of people across the world actively use social media today. This abrupt societal transition has led to a dramatic increase in the extent to which a person’s social footprint is documented online in the public domain. While for some social media sites, like Snapchat and Facebook, privacy is a key feature, on sites like Twitter, comments are intentionally made public for the world to see. The ever growing number of intentionally public interactions creates new opportunities for organizations to better understand consumers and how they feel about specific issues or products. In this chapter we will discuss social polling and influencer analytics, which are two of the most popular use cases for Social Media Analytics. We will also highlight an emerging trend across multiple industries where organizations are using aggregate social polling as input to demand forecasting solutions. Data for social analytics is largely unstructured and the social graph is massive. As a result, the choice of analytics techniques can have an enormous impact on the quality of the results and ROI for businesses that undergo analytics initiatives. As such, we will cover and discuss the relative merits of a variety of popular analytics techniques, across industry and academia, addressing best practices for these use cases.