Exploiting entity-level morphology to Chinese nested named entity recognition

Named entity recognition plays an important role in many natural language processing applications. While considerable attention has been pain in the past to research issues related to named entity recognition, few studies have been reported on the recognition of nested named entities. This paper presents a morpheme-based due-layer labeling method to Chinese nested named entity recognition. To approach this task, we first employ the logistic regression model to extract multi-level entity morphemes from an entity-tagged corpus, and thus explore multiple features, particularly entity-level morphological cues for Chinese nested named entity recognition under the framework of conditional random fields. Our experimental results on different datasets show that our system is effective for most nested named entities under evaluation, illustrating the benefits of using entity-level morphology.