Chinese shallow semantic parsing based on multilevel linguistic clues

With rapid development of artificial intelligence and Chinese information processing technology, research related to natural language processing have reached the level of semantic understanding gradually, while Chinese Shallow Semantic Parsing is the key technique in the semantic understanding field. In this paper, a further improvement is conducted on the basic model of Chinese semantic role labeling for linear classification based on conditional random fields. In this paper, a method of combination of linguistic clues, combining with the existing linear sequence labeling algorithm and integrating some multilevel linguistic clues, such as morphology is related to syntax, in the model training to reconstruct and improve the Chinese semantic role labeling model of linear sequence. Through the experimental comparison and linguistic assistant analysis, this paper puts forward a targeted improvement method to significantly improve the accuracy of model labeling and proves that the integration of related linguistic clues in the semantic role labeling model based on linear sequence can improve the effect of model labeling.