alatenttopicmodelforcompleteentityresolution-Resolution.PDFVIP

A Latent Topic Model for Complete Entity Resolution Liangcai Shu, Bo Long, Weiyi Meng Department of Computer Science, SUNY at Binghamton Binghamton, New York 13902, U. S. A. lshu@ blong1@ meng@ Abstract— In bibliographies like DBLP and Citeseer, there are three kinds of entity-name problems that need to be solved. First, multiple entities share one name, which is called the name sharing problem. Second, one entity has different names, which is called the name variant problem. Third, multiple entities share multiple names, which is called the name mixing problem. We aim to solve these problems based on one model in this paper. We call this task complete entity resolution. Different from previous work, our work use global information based on data with two Fig. 1. An example of the name sharing problem (left) and an example of types of information, words and author names.Weproposea name variant problem (right) in DBLP. generative latent topic model that involves both author names and words — the LDA-dual model, by extending the LDA (Latent Dirichlet Allocation) model. We also propose a method to obtain model parameters that is global information. Based on obtained disambiguation [2][3], object distinction [4], mixed citation model parameters, we propose two algorithms to solve the three [5] and author disambiguation [6]. problems mentioned above. Experimental results demonstrate The name variant problem the effectiveness and great potential of the proposed model and The second problem is called the name variant problem algorithms. or one-entity-multiple-name problem. An author may have