Automatic Entity Relation Extraction Based on Conditional Random Fields

Entity relation extraction (RE) is one of the important research fields in information extraction, we regard RE as a classification problem in this paper. This paper presents a novel approach, conditional random fields (CRFs)-based machine learning is used to extract entity relation between entities from Chinese texts, ten features have been designed for entity relation extraction, which includes morphology, grammar and semantic feature. Experimental results show that the new approach achieve an improved performance on corpus for Chinese entity relation extraction.