Chinese Natural Chunk Research Based on Natural Annotations in Massive Scale Corpora

Great changes in Natural Language Processing (NLP) research appear with the rapid inflation of corpora scale. NLP based on massive scale natural annotations has become a new research hotspot. We summarized the state of art in NLP based on massive scale natural annotated resource, and proposed a new concept of “Natural Chunk”. In the paper, we analyzed its concept and properties, and conducted experiments on natural chunk recognition, which exhibit the feasibility of natural chunk recognition based on natural annotations. Chinese natural chunk research, as a new research direction in language boundary recognition, has positive influences in Chinese computing and promising future.