Semantic Parsing for Single-Relation Question Answering

Citations Over TimeTop 1% of 2014 papers

Abstract

We develop a semantic parsing framework based on semantic similarity for open domain question answering (QA). We focus on single-relation questions and decompose each question into an entity mention and a relation pattern. Using convolutional neural network models, we measure the similarity of entity mentions with entities in the knowledge base (KB) and the similarity of relation patterns and relations in the KB. We score relational triples in the KB using these measures and select the top scoring relational triple to answer the question. When evaluated on an open-domain QA task, our method achieves higher precision across different recall points compared to the previous approach, and can improve F 1 by 7 points.

Related Papers

• → Building A Japanese Parsed Corpus(2003)33 cited
• → Discourse Representation Structure Parsing for Chinese(2023)1 cited
• Research on Auto-identifying of Adjoining Relation Markers Based on Rule(2012)
• An Empirical Investigation of Error Types in Vietnamese Parsing(2018)
• Towards Understanding and Applying Chinese Parsing using Cparser(2020)