Enhancing collective entity resolution utilizing Quasi-Clique similarity measure

Entity resolution(ER) is the problem of identifying duplicate references that refer to the same real world entity. It is a critical component of data integration and data cleaning. Attribute-based entity resolution is the traditional approach where similarity is computed for each pair of references based on their attributes. More recently, context-base entity resolution has been proposed which considers the attributes of the related references. In this paper, we present a collective entity resolution approach which using quasi-clique similarity to improve the accuracy. It complements the traditional methodology by reducing the number of false positive. An experimental evaluation on several datasets shows high recall and precision rates, which validate the method´s efficiency.