Skip to main content
Open Access Publications from the University of California


UCLA Electronic Theses and Dissertations bannerUCLA

Using Social Graph Data to Enhance Expert Selection and News Prediction Performance


Human intuition leads us to believe in the existence of experts, individuals with knowledge or insight that exceeds that of an average person. Can the idea of experts be harnessed to accurately perform popular news prediction? Can they perform this task better than "the crowd", a collection of all or large amounts of the entire population? We explore this concept, first introducing various expert selection strategies, and then attempting to improve on them through the use of social graph data. We also examine the possibility of using expert characteristics and social data as parameters for machine learning models. Ultimately, we make two conclusions: it is extremely difficult for expert wisdom to outperform crowd wisdom, but expert selection can be used as a means of resource efficient sampling.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View