Patent prior art search uses dispersed informationto retrieve all the relevant documents with strong ambiguityfrom the massive patent database. This challenging task consists in patent reduction and patent expansion. Existing studies on patent reduction ignore the relevance between technical characteristics and technical domains, and result in ambiguous queries. Works on patent expansion expand termsfrom external resource by selecting words with similar distribution or similar semantics. However, this splits the relevance between the distribution and semantics of the terms.Besides, common repository hardly meets the requirementof patent expansion for uncommon semantics and unusualterms. In order to solve these problems, we first present anovel composite-domain perspective model which convertsthe technical characteristic of a query patent to a specificcomposite classified domain and generates aspect queries.We then implement patent expansion with double consistencyby combining distribution and semantics simultaneously. Wealso propose to train semantic vector spaces via word embedding under the specific classified domains, so as to provide domain-aware expanded resource. Finally, multiple retrieval results of the same topic are mergedbased on perspective weight and rank in the results. Our experimental resultson CLEP-IP 2010 demonstrate that our method is very effective. It reaches about 5.43%improvementin recall and nearly12.38% improvement in PRES over the state-of-the-art. Ourwork also achieves the best performance balance in terms ofrecall, MAP and PRES.
详细信息：https://journal.hep.com.cn/fcs/EN/10.1007/s11704-018-7056-6
Patent expanded retrieval via word embedding under composite-domain perspectives.pdf