Finding Experts in Social Media Data using a Hybrid Approach

TL;DR

This paper presents a hybrid expert finding system called ExpertQuest that combines content analysis, social graph analysis, and Semantic Web technologies to improve expert identification in social media data.

Contribution

It introduces a novel integrated approach and a prototype system that leverages multiple techniques for expert finding in social media and Linked Data.

Findings

Prototype system ExpertQuest developed and evaluated.

Hybrid approach shows practical benefits and limitations.

Uses modern functional programming language (Clojure).

Abstract

Several approaches to the problem of expert finding have emerged in computer science research. In this work, three of these approaches - content analysis, social graph analysis and the use of Semantic Web technologies are examined. An integrated set of system requirements is then developed that uses all three approaches in one hybrid approach. To show the practicality of this hybrid approach, a usable prototype expert finding system called ExpertQuest is developed using a modern functional programming language (Clojure) to query social media data and Linked Data. This system is evaluated and discussed. Finally, a discussion and conclusions are presented which describe the benefits and shortcomings of the hybrid approach and the technologies used in this work.