# KDD Papers

### Automatic Synonym Discovery with Knowledge Bases

Meng Qu (University of Illinois at Urbana-Champaign);Xiang Ren (University of Illinois at Urbana-Champaign);Jiawei Han (University of Illinois at Urbana-Champaign)

#### Abstract

Recognizing entity synonyms from text has become a crucial task in many entity-leveraging applications. However, discovering entity synonyms from domain-specific text corpora (\textit{e.g.}, news articles, scientific papers) are rather challenging. Current systems take an entity name string as input to find out other names that are synonymous, ignoring the fact that often times a name string can refer to multiple entities (\textit{e.g.}, apple’’ could refer to both \textit{Apple Inc} and the fruit \textit{apple}). Moreover, most existing methods require training data manually created by domain experts to construct supervised-learning systems. In this paper, we study the problem of automatic synonym discovery with knowledge bases, that is, identifying synonyms for \textit{knowledge base entities} in a given domain-specific corpus. The manually-curated synonyms for each entity stored in a knowledge base not only form a set of name strings to \textit{disambiguate} the meaning for each other, but also can serve as \textit{distant}’’ supervision to help determine important features for the task. We propose a novel framework, called \textbf{DPE}, to integrate two kinds of mutually-complementing signals for synonym discovery, \textit{i.e.}, \textit{distributional features} based on corpus-level statistics and \textit{textual patterns} based on local contexts. In particular, DPE jointly optimizes the two kinds of signals in conjunction with distant supervision, so that they can mutually enhance each other in the training stage. At the inference stage, both signals will be utilized to discover synonyms for the given entities. Experimental results prove the effectiveness of the proposed framework.