This paper proposes a novel two-stage method for bilingual product name dictionary construction from comparable corpora. In previous work, some researchers s-tudy the problem of expanding a set of given seed entities into a more complete set by discovering other entities that also belong to the same concept, it just solves the problem about expansion of entity set in a monolingual language, but the expansion of bilingual entity is really blank problem from comparable corpora. A typical example is to use "Honda-本田"as seed entity, and derive other entities(e.g., "Ford-福特") in the same concept set of product name. We address this problem by utilizing a two-stage approach based on entity set expansion and bilingual entity alignment from comparable corpora. Evaluations using English and Chinese reviewer corpus verify that our method outperforms conventional methods.
展开▼