Microsegmenting Users On Twitter Using Hierarchical Interest Graph

Introduction

Determining user interests from the content they generate on social networking platforms, such as Twitter, has gained a lot of attention due to the explosive usage of social networks. However, the subsequent problem of micro-segmenting users into appropriate categories based on their interests is not well addressed. In this paper, we present an approach that leverages the category graph of wikipedia and an adaptation of the spreading activation algorithm to abstract user interests into appropriate levels of categories. Although, the main motivation of this work is to categorize users into buckets (categories) for business intelligence, we also discuss the advantages of the approach for personalization and recommendation systems.