An optimization framework for entity recognition and disambiguation.

ABSTRACTWe present a system for entity recognition and disambiguation (ERD) in short text, aiming at identifying all text fragments referring to an entity contained in Freebase. The task is organized in two steps. Given a short text the first step is discovering text fragments which possibly refer to an entity.Since multiple entities may share common mention, identifying which entity the mention is referring to in the given short text is necessary. Our system integrates three kinds of features: mention-entity similarity, entity-entity similarity and context-mention entity similarity. By considering every possible combination of mention-entity pair, we select the one with highest confidence score. An implementation of our system is described, along with our evaluation results. Experiments show that the proposed features improve the performance to a certain extent.