Research And Application Of The Approach To Understand Elementary Mathematics Based On Sentence Framework And Fuzzy Matching

With the continuous development of artificial intelligence technology,it has been gradually penetrated into various fields.How to apply artificial intelligence to the education industry has become a very hot research direction.It has been well developed on the issues of elementary mathematics meaning understanding,auto-solving,and automatic proof in recent years.As one of the most basic tasks in artificial intelligence,new methods of natural language processing have also emerged in large numbers,and the accuracy of related tasks has also been continuously improved.This paper implements an elementary mathematics meaning understanding system based on the related technology of natural language processing by using the knowledge of sentence pattern matching and semantic similarity,and tests the functional module of the system and analyzes the result of the test in depth.This paper concentrates on the following researches:1)Elementary mathematics representationFirstly,this paper discusses how to express elementary mathematics knowledge in computers,lists the current mainstream knowledge representation methods and analyzes the applicability of various methods.On this basis,this paper uses object-oriented method to express knowledge,expresses every relationship in elementary mathematics as a class in object-oriented language,and represents the relationship between relationships by using its inheritance and encapsulation.2)Named Entity Recognition Based on CRF++Secondly,this paper discusses the issues related to Chinese word segmentation,part of speech tagging,and Chinese named entity recognition in natural language processing and introduces the principle of conditional random fields and how to identify named entities by using tool CRF++ training model.Through the testing and analysis of the model,the accuracy of the model in identifying the relevant entities in elementary mathematics reached 95%.3)Research on Sentence Pattern Based on Fuzzy Matching AlgorithmIn the next step,this paper introduces related theoretical knowledge about sentence models,relevant algorithms of sentence pattern matching,and how to apply sentence patterns to the understanding of elementary mathematics.And the fuzzy matchingmethod is proposed directing at some limitations of sentence pattern matching before which this paper introduces a new method based on word2 vec by analyzing some methods of transforming texts to vectors.Also,this article describes the related principles of word2 vec and its two different models,and analyzes the advantages of this method.4)Implementation of Elementary Mathematics Problem Understanding System Based on Sentence Pattern and Fuzzy MatchingLastly,the paper constructs an elementary mathematics problem intention understanding system in the combine of the relevant knowledge of sentence pattern and fuzzy matching,laying the foundation of the automatic solving of elementary mathematics.The method is proved correct through the test and analysis of the system,the accuracy of understanding of elementary mathematics problems is beyond 80%.