A system and application method for automatic question answering by a reading robot

Abstract

The invention discloses a method and system for automatic question-and-answer by a reading robot, including a question type analysis module, a word similarity matching module, a paragraph topic sentence and an author's opinion sentence extraction module, and an answer sentence extraction module. question features, and then perform linear interpolation to analyze question types, which effectively fills up the blanks of only analyzing question types based on question sentences; the present invention can extract some paragraph topic sentences and author opinion sentences that are different from question keywords and frames The matching answer candidate sentences are recalled; the present invention can realize the intelligent answering of questions and answers in reading comprehension, effectively improving the answering rate of automatic questions and answers; the method proposed by the present invention can also be applied to natural problems such as text implication, automatic summarization, and complex problem solving. in language processing tasks.

Description

A system for automatic question answering by reading robot and its application method technical field The present invention relates to natural language processing field, relate in particular to the method and system that reading robot carries out automatic question and answer system. Background technique [0002] In recent years, the research on question answering technology has received extensive attention. Start developed by the Artificial Intelligence Laboratory of the University of Technology, AnswerBus of the University of Michigan, Siri and WeChat developed by Apple Cotana developed by Soft Company. There are also some robots centered on intelligent question-answering technology in China, such as: Microsoft's small Bing, Baidu's Xiaodu, etc., but the question and answer research based on reading comprehension is far from enough. [0003] At present, there are mainly two research directions based on retrieval model and based on deep learning. Based on traditiona...

AI Technical Summary

Problems solved by technology

[0004] At present, the method based on retrieval model and deep learning has achieved good results on general reading comprehension problems, but the effect on Chinese reading comprehension questions in the college entrance examination is not satisfactory. This is because the reading comprehension problems in the college entrance examination are more complicated. Requires deep semantic understanding and reasoning about questions and background material

In addition, although deep learning-based methods can automatically learn features, they require sufficient data for training.

Images