Joint Link-Attribute User Identity Resolution in Online Social Networks
Sergey Bartunov, Korshunov Anton, Seung-Taek Park, Ryu Wonho, Hyungdong Lee · 2012
In the modern Web, it is common for an active person to have several proles in dierent online social networks. As new general-purpose and niche social network services arise every year, the problem of social data integration will likely remain actual in the nearest future. Discovering multiple proles of a single person across dierent social networks allows to merge all user’s contacts from dierent social services or compose more complete social graph that is helpful in many social-powered applications. In this paper we propose a new approach for user prole matching based on Conditional Random Fields that extensively combines usage of prole attributes and social linkage. It is extremely suitable for cases when prole data is poor, incomplete or hidden due to privacy settings. Evaluation on Twitter and Facebook sample datasets showed that our solution signicantly outperforms common attribute-based approach and is able to nd matches that are not discoverable by using only prole information. We also demonstrate the importance of social links for identity resolution task and show that certain proles can be matched based only on social relationships between online social networks users.