UC San Diego

Searching for people on the web is a ubiquitous activity. We search for people on web search engines, social networks, publication websites and various other websites. Although the quality of web search has improved over the decade, finding information regarding people has become increasingly difficult. One of the major issues relating to it is with regards to different people having the same name. Name disambiguation can occur when trying to find person in a large web space. We have build an efficient integrative framework to solve the problem of name disambiguation across not just one but several domains. The thesis presents an algorithm to aggregate and cluster people from various domains using their rich feature space of biographic facts and their connections. We demonstrate the effectiveness of our approach by testing the efficacy of the disambiguation algorithm and its impact on person search