Year of Award

2014

Document Type

Thesis

Degree Type

Master of Science (MS)

Degree Name

Computer Science

Department or School/College

Department of Computer Science

Committee Chair

Jesse Johnson

Commitee Members

Robert Smith, Douglas Dalenberg

Keywords

recommender systems, link prediction, social networks, machine learning, graph traversal, graph database

Subject Categories

Computer Sciences

Abstract

The paper examines the link prediction problem for directed multi-modal complex networks. Specically, a hybrid method combining collaborative filtering and Triadic Closeness methods is developed. The methods are applied to a sample of the GitHub network. Implementation details are discussed, with a focus on design of a scalable system for handilng large data sets. Finally, results of this new method are discussed with no significant improvement over current methods.

Share

COinS
 

© Copyright 2014 William Lyon