A method for calculating the similarity of questions containing personal names in the field of question answering

Abstract

The invention discloses a method for calculating the similarity of question sentences containing personal names in the question-and-answer field, which calculates the similarity of personal names and non-personal names respectively, and calculates the similarity of sentences from the word order and length of sentences in consideration of the sentence structure , and finally obtain the similarity of the whole sentence according to the weighting of sentence semantic similarity and structural similarity. It solves the problem that human subjective judgment cannot be obtained when calculating sentences containing personal names through the previous sentence similarity algorithm. The method proposed by the invention can more accurately calculate the similarity of sentences containing personal names, and can be widely applied to the field of question answering.

Description

technical field [0001] The invention relates to the technical field of question answering, in particular to a method for calculating the similarity of question sentences containing personal names in the field of question answering. Background technique [0002] The similarity calculation of questions has always been the basic and important research work in the field of artificial intelligence and natural language processing. It is also a research hotspot and has a very wide range of applications, such as question answering systems and information retrieval systems. [0003] At present, the algorithms for calculating the similarity of Chinese sentences can be roughly divided into the following categories: the first category is based on feature words; the second category is based on sentence structure; the third category is based on semantic dictionaries. First of all, the first type of method is based on the method of feature words. This method refers to extracting their feat...

AI Technical Summary

Problems solved by technology

[0005] The present invention aims to solve the problem that the difference between names and the importance of names to the entire sentence cannot be reflected when calculating the similarity of sentences in the current question-and-answer field, which will lead to unsatisfactory results of the similarity calculation of questions. It provides a A question-sentence similarity calculation method containing personal names in the question-and-answer field to improve the accuracy of question-sentence similarity calculations

Images