Dissertation > Excellent graduate degree dissertation topics show

Ontology -based Knowledge Base Answering System Dynamic Research and Implementation

Author: MaLi
Tutor: TangSuQin
School: Guangxi Normal University
Course: Computer Software and Theory
Keywords: Answering System Questions deal Ontology Dynamic ontology Formal Concept Analysis Automotive business area
CLC: TP391.6
Type: Master's thesis
Year: 2009
Downloads: 138
Quote: 6
Read: Download Dissertation

Abstract

With the rapid development of Internet, the network grows exponentially the amount of information, the information to answer user questions become important information sources. Traditional search engines information retrieval approach, while a certain extent, the information on the user query, but it returns a large number of related or unrelated to link to this page, the user is difficult from these links to quickly find the answers they wanted. Therefore, it is a genuine desire to meet their needs, intelligent information retrieval methods. QAS (Question Answering System, abbreviated QA) providing people with a natural language man-machine interface, this approach is consistent with the expressions that people question the other hand, with respect to the way the search engine keywords in understanding questions intent question has obvious advantages. And answering system eventually returned to the user is the most relevant answers, compared to traditional search engines better performance. Among the many types of question answering system, answering system based on ontology knowledge base has to answer questions accurately and quickly, and can be to a certain extent reasoning, etc., to meet people for intelligent, fast and accurate information retrieval needs, so research is based on dynamic ontology knowledge quiz system has a very important theoretical and practical significance. This paper studies the dynamic ontology knowledge quiz based system construction and implementation. First studied the technical characteristics of ontology and knowledge base, and then design an ontology knowledge quiz based on dynamic system model; followed by analysis of the various questions processing method, an improved understanding of the sentence mode matching questions algorithm, and using this algorithm semantic comprehension questions; Then in order to overcome the current question answering system based on ontology knowledge base with knowledge bottleneck problem of insufficient scale to study a dynamic ontology construction method, which can be as Web information changes automatically update the contents of the body, which can improve ontology-based knowledge quiz system performance; Finally, build the ontology knowledge quiz based on dynamic system model to achieve a quiz for automotive systems business area. This study mainly includes the following sections: (1) ontology knowledge quiz based on dynamic model of the system design. Answering system based on ontology knowledge base and the main drawback Research and combine the needs of the project, designed by the front and back of natural language interfaces to build dynamic ontology knowledge base composed of two main modules ontology knowledge quiz based on dynamic system model. (2) Natural Language Understanding HCI research. Handle the user's natural language question answering questions is a priority, but also a crucial task. Relies entirely on natural language processing technology questions processing method is still not mature enough to understand questions can not achieve the desired effect, this paper designs an improved understanding of sentence mode-matching questions algorithm that does not require complex syntax analysis, semantic analysis process, through the sentence surface morphology can understand the semantics of questions, tested the algorithm is compared with the existing sentence comprehension mode-matching method in the questions the accuracy and success rate has improved significantly in the actual application has achieved very significant results. (3) Construction of a dynamic ontology knowledge base. Ontology is currently a hot research. The traditional static text can not meet people's growing demand for knowledge, rich Web information has become a new source of knowledge acquisition. In this paper, ontology current research and projects need to study a dynamic ontology construction method DOCM. The method in order to reduce the burden of knowledge engineers, using semi-automatic way to build the body, and the method can also be used formal concept analysis Web information technology will be automatically added to the body, for the body to be dynamically updated. In order to verify the feasibility of the method, the paper also describes the dynamic field of automotive business ontology building process. (4) the field of automotive business answering system implementation. This article is in Guangxi, scientific research and technology development projects - manufacturing-oriented e-commerce platform based on active search for the study. As people's living standards continue to improve and manufacturing technology continues to evolve, more and more people began to enter the ranks of the car, but also makes people information in the field of automobile business increased demand. Since there is no one for the automotive business fields answering system, so this dynamic ontology-based knowledge base answering system based on the study design and implementation of an automotive business area quiz system that can answer user substantially all of the field Most questions, has a strong practical value. In addition to introducing the system implementation process, but also the realization of the system is given interface and performance analysis, the experimental results demonstrate the feasibility of the theoretical study. In the last part of the work has been completed on the conclusion and the need to further improve the work are described in detail.

Related Dissertations

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Teaching machine, learning machine
© 2012 www.DissertationTopic.Net Mobile