Semantic association and matching method of Chinese natural language conversations

Abstract

The invention discloses a semantic association and matching method of Chinese natural language conversations. According to the method, word segmentation and sentence segmentation are carried out on Chinese conversation sentences collected at a time; an input sentence word segmentation data structure is created according to each word and sentence segmentation result and the intensity relation dataand activation relation data conditions of segmented words in a word library of a database corresponding to the segmented words; the accurate positions of verb keywords, prepositive keywords and postpositive keywords of each segmented sentence in the input sentence word segmentation data structure are determined; primary matching screening is carried out on sentence records in a notice informationbase by means of the created input sentence word segmentation data structure, multiple sentence records are obtained after matching screening, the semantic confidence degree of each sentence record is determined, comparison is conducted on the semantic confidence degree of each sentence record, the sentence record with the highest semantic confidence degree is selected as the optimal semantic matching sentence and output as a result; finally, input sentence character string information is stored into the notice information base according to a corresponding format to serve as a new sentence record in the notice information base.

Description

technical field [0001] The invention belongs to the technical field of natural language processing, and in particular relates to a semantic association and matching method for Chinese natural language dialogue. Background technique [0002] At present, people's language cognition and natural language semantic understanding are still in the exploratory stage. The current semantic relational information retrieval technology that has been oriented to practical applications basically only stays on the shallow method of keyword matching, and the range of search results is quite wide, requiring people to find useful information from them again. The most typical examples are: Internet web search engines; on the other hand, in the field of academic research, most of the current natural language understanding technologies use probabilistic and statistical models to analyze and process text, and there are many researches on sentence segmentation, large-scale corpus annotation and cons...

AI Technical Summary

Problems solved by technology

The current semantic relational information retrieval technology that has been oriented to practical applications basically only stays on the shallow method of keyword matching, and the range of search results is quite wide, requiring people to find useful information from them again. The most typical examples are: Internet web search engines; on the other hand, in the field of academic research, most of the current natural language understanding technologies use probabilistic and statistical models to analyze and process text, and there are many researches on sentence segmentation, large-scale corpus annotation and construction, and speech recognition. Most of the technologies for semantic processing of Chinese sentences stay at the stage of theoretical research, and there are few applications that can solve practical problems in real life

Images