REMI: Mining Intuitive Referring Expressions on Knowledge Bases
Abstract
A referring expression (RE) is a description that identifies a set of instances unambiguously. Mining REs from data finds applications in natural language generation, algorithmic journalism, and data maintenance. Since there may exist multiple REs for a given set of entities, it is common to focus on the most concise and informative (i.e., intuitive) ones. We present REMI, a method to mine intuitive REs on large knowledge bases. Our experimental evaluation shows that REMI finds REs deemed intuitive by users. Moreover we show that REMI is several orders of magnitude faster than an approach based on inductive logic programming.
Origin : Publisher files allowed on an open archive