Research And Application Of A Stratified Sampling Problems Automatic Understanding Method Merging Entity Semantic Relations

In recent years,with the rapid development of artificial intelligence,all aspects of life have become more and more intelligent.In the field of education,it is worth discovering how artificial intelligence can be applied to better support education.One of the prerequisites to establish the personalized computer tutor for students is to achieve computer automatic problem solving.At the same time,the automatic mathematical problems solving has always been an extremely challenging and attractive research area in the field of artificial intelligence,and the first challenge is to realize achieve the computer automatic problem understanding.The stratified sampling problem,as a fixed knowledge point for the college entrance examination over the years,has always been a difficult problem in the automatic understanding because of its short text and strong situational characteristics.At present,there is relatively little research on the automatic stratified sampling problems understanding.This study takes stratified sampling problems as the research object and designs a method for automatic understanding of these questions by combining entity semantic relations through multi-angle analysis of stratified sample questions and question intentional features.And various experiments have been performed to prove the effectiveness of this method.At the same time,a stratified sampling problems automatic understanding system has been developed implemented with the proposed method.First of all,in the study of the stratified sample problem sentences and problems understanding,through the analysis of the characteristics of the stratified sample problem sentences,the difficulty in understanding the stratified sample problems and the direction of problem understanding method is clarified.At the same time,through the research on the stratified sample problems understanding,a set of stratified sampling problem understanding framework was defined,which indicated the objectives and the core process of problem understanding.Secondly,an understanding method of stratified sample problems merging entity semantic relations is designed on the basis of the research on stratified sample sentences and problem understanding.The validity of this method was verified by the test corpora of the college entrance examination problems and the stratified sampling problems on the internet.The experimental results show that:in addition to the overall sentence,the F value of this method for all types of stratified sample sentences has reached more than 90%,the recognition rate of the overall sentence has reached 80.7%.This method can extract the key parameters effectively,and the whole problems recognition rate reached 76%.The data showed that the understanding method of stratified sample problems merging entity semantic relations obtained a satisfied effect.In the end,a stratified sample problem understanding system was designed based on the understanding method of stratified sample problems merging entity semantic relations.It has been tested that the system can better understand the stratified sample problems of simple and compound sentence patterns,which meets the basic needs of automatic understanding of stratified sampling problems.