Chinese notional word extraction algorithm based on semantic comprehension

Abstract

The invention provides a Chinese notional word extraction algorithm based on semantic comprehension. The method includes the steps: building a data retrieval structure by a hash tree dictionary; segmenting a Chinese sentence into short sentences according to a segmentation table, and storing character string matching information in a matching process when participles are matched; judging an existing ambiguous field by the character string matching information and word-by-word scanning, and processing pre-segmented intermediate results in a segmenting process. The Chinese notional word extraction algorithm based on semantic comprehension improves the ambiguous field processing course of the participles, and has more ideal time complexity and segmentation accuracy.

Description

technical field [0001] The invention relates to natural language processing, in particular to a Chinese content word extraction algorithm based on semantic understanding. Background technique [0002] With the gradual maturity of network technology and the Internet, the traditional single keyword method can no longer meet the current content acquisition needs of massive information. How to design a question answering system has become an important technology that needs to be solved in network search. From the point of view of the existing question answering system, due to the complexity of word segmentation and the limitations of semantic recognition in Chinese, the technology of formed products is relatively backward. For example, because the existing word segmentation method must first set an initial value of matching word length, If the word length is too long, the time complexity of the algorithm will increase; if the word length is too short, the accuracy of segmentatio...

AI Technical Summary

Problems solved by technology

[0002] With the gradual maturity of network technology and the Internet, the traditional single keyword method can no longer meet the current content acquisition needs of massive information. How to design a good question answering system has become an important technology that needs to be solved in network search.

From the perspective of the existing question answering system, due to the complexity of word segmentation and the limitations of semantic recognition in Chinese, the technology of formed products is relatively backward. For example, because the existing word segmentation method must first set an initial value of matching word length, If the word length is too long, the time complexity of the algorithm will increase; if the word length is too short, the accuracy of segmentation will decrease

The processing of ambiguous fields cannot meet the needs of actual users

Images