Research On Automatic Question Answering System Based On Ontology

With the popularization of the Internet, education through networks is becomingincreasing popular. Network teaching gradually becomes an effective way for people’sstudying. Because of its own limitations of network teaching, such as time, environment andgeographical, the question answering system becomes a useful complement. Automaticquestion answering system, also known as auto-answering system can be used as studentsanswering system of network teaching. Automatic question answering system is a hotresearch issue in the field of natural language processing. With the development of artificialintelligence, different modes of implementation have emerged for automatic questionanswering system. Ontology has a good concept hierarchical structure and the support oflogical reasoning, with the ability to express semantics through the relationship among theconcepts. With the deep research on the theory and application of ontology, an increasingnumber of research institutes and academics try to apply ontology to information retrievaland automatic question answering system. Ontology-based Automatic Question AnsweringSystem, as a specific application of the study, has attracted more and more researchers’attention.This paper will go for a further research base on the results of previous studies, themain contents of paper includes: this paper first provides an overview of the developmentstatus of automatic question answering system, and then introduce the basic concepts ofontology, focus on the ontology application in automatic question answering system, nextelaborate the key technologies of natural language processing in automatic questionanswering system (such as segment, stop words processing and similarity calculating),including word similarity calculating and sentence similarity calculating. As for existingshortcomings of the existing automatic question answering systems, this paper proposessome improvements, such as the segmentation algorithm, words similarity calculatingalgorithms and sentences similarity calculating algorithm. Finally, take “Data Structure”curriculum ontology for example, compare to the existing question answering systemsthrough experiments to verify the performance of the automatic question answering systemproposed this paper. We can learn from the experimental results of this article automaticquestion answering system in this paper can improve recall rate and precision rate ofautomatic question answering system to some extent.