Application Research of Similarity Algorithm in the Design of English Intelligent Question Answering System

Abstract

Question answering systems can answer user questions accurately and instantly, and are gradually becoming the focus and hotspot of scholars' research. In various question answering systems, the key to determining the accuracy of question answering is how to find similar sentences, so the study of sentence similarity algorithm has become a very basic and important task. The purpose of this article is to study the application of similarity algorithm in the design of intelligent English question answering system. It first introduces the related technologies of intelligent question answering system, and then mainly introduces the calculation method of sentence similarity based on WordNet. Among them, the WordNet semantic dictionary is used to calculate the semantic information of the sentence, find the longest word matching method in the sentence, and then apply the WordNet sentence similarity algorithm to the automatic question answering system. This new method is used for the design of the intelligent English question answering system with realization. Finally, the system performance test is carried out, 500 repeated sentences are divided into 5 groups, and the correct rate of each test set group is finally calculated. The results show that the average accuracy of the entire answering system is accurate, and the percentage is above 75%, which has shown an excellent level.