A compositional-distributional semantic model for searching complex entity categories

Abstract

Users combine attributes and types to describe and classify entities into categories. These categories are fundamental for organising knowledge in a decentralised way acting as tags and predicates. When searching for entities, categories frequently describes the search query. Considering that users do not know in which terms the categories are expressed, they might query the same concept by a paraphrase. While some categories are composed of simple expressions (e.g. Presidents of Ireland), others have more complex compositional patterns (e.g. French Senators Of The Second Empire). This work proposes a hybrid semantic model based on syntactic analysis, distributional semantics and named entity recognition to recognise paraphrases of entity categories. Our results show that the proposed model outperformed the comparative baseline, in terms of recall and mean reciprocal rank, thus being suitable for addressing the vocabulary gap between user queries and entity categories.