Dissertation > Excellent graduate degree dissertation topics show

Domain knowledge uncertainty reasoning

Author: HuangFu
Tutor: LiYong
School: Kunming University of Science and Technology
Course: Pattern Recognition and Intelligent Systems
Keywords: Domain knowledge Noumenon Bayesian networks Conditional Event Algebra Uncertainty reasoning
CLC: TP391.1
Type: Master's thesis
Year: 2011
Downloads: 57
Quote: 0
Read: Download Dissertation

Abstract


Knowledge as a new area of ??current research in the field of individual cognition can not only significantly enhanced working memory capacity, but also improve the predictive ability of the information. Domain ontology (Ontology) as an important model of knowledge modeling tool, it can not only described in the semantic and knowledge level of knowledge, and the standardization of related concepts, and clear depiction. So as to lay a foundation for the sharing of knowledge, domain knowledge of the organization and build a good platform. However, the intersection between the degree of expression of the concept of the body is not well, can not express reasoning mode concept only know part of the information, and that uncertainty. Therefore, uncertainty exists in the domain knowledge of the body structure, has become the focus of attention and research. Bayesian decision theory to prepare a strong theoretical basis for the treatment of various uncertain events or reasoning. In the expression of the uncertainty of the domain knowledge and reasoning, Bayesian networks are a wide range of applications, and has proven to be one of the most effective methods of uncertainty knowledge confidence. Conditional event algebra as an emerging discipline in dealing with uncertainty, probability and fuzzy reasoning problems have strong mathematical and theoretical foundation. Conditional event algebra to higher-order conditional event converted into ordinary events and logical combination of events. In this paper, Bayesian networks and conditional event algebra uncertain events or knowledge exist in the organization of the domain knowledge related to the study. This paper first introduces ontology construction technology. Domain ontology as the carrier of the field of knowledge, able to provide intuitive common understanding of the field of knowledge, vocabulary clear identity within the field, and give these vocabulary (terminology) and a clear definition of the relationship between the vocabulary. The introduction of domain ontology building process and its description language. The probability of extended ontology, it has the ability to indicate uncertainty information. Eventually build a domain ontology with probability information. Second, uncertainty reasoning based on Bayesian network. The Bayesian network conditions of the assumption of independence between random variables, a joint probability distribution of the image will be represented as a graph structure and a series of conditional probability tables, corresponding elimination of variables, a variable calculated rectifiable served probability distribution or part of the variable probability distribution. Bayesian networks are widely used for uncertainty reasoning, one of the most important reason is the probability theory is a reasonable way to represent uncertain. In this paper, Stanford University ontology development tools protege3.3.1 of building domain ontology, domain ontology based on the probability of extended use the Jena and related components to parse the probability of extended body. Then parse the ontology file for formatting, further complete the construction of the Bayesian network. Graphical way through the building of Bayesian network domain knowledge uncertainty reasoning process. Finally, the uncertainty reasoning based on conditional event algebra combined Bayesian network. Conditional event algebra as an emerging discipline, with a wide range of applications in reasoning uncertain information. Conditional event algebra and conditions of the event, by extending normal measurable space, so that the probability of logical expression rules consistent higher order conditional event to become a regular event and the logical combination of events and conditional event algebra. The final completion of the reasoning process of high-level events. This paper demonstrated the feasibility and effectiveness of the method in solving the uncertain information in the field of knowledge.

Related Dissertations

  1. Yellow phosphorus storage tank area of Safety Evaluation,TQ126.317
  2. Applied Research of Ontology in Intelligent Residence Community,TP391.1
  3. Software based on Bayesian network model and implementation of risk management,TP311.52
  4. Based on Bayesian Networks motor fault diagnosis method,TM307.1
  5. Study on Chinese Text Classification Combined with Ontology,TP391.1
  6. Research on Ontology Modeling and Integration for Collaborative Design,TP311.52
  7. Research on Hydrological Risk Management during Operation of the Middle Route Project for South-to-North Water Transfer,TV12
  8. Research and Application on Heterogeneous Data Integration of Smart Grid,TM76
  9. Water Quality Pollution Evaluation and Prediction Based on Bayesian Networks,X824
  10. The Application of Dynamic Bayesian Networks in Hydrologic Forecast,TV124
  11. Recommended model based on Bayesian network Senate elective system applied research,G647
  12. Research on Analysis and Prediction Model of Road Traffic Accident by Data Mining,U491.31
  13. The Design and Implementation of Semantic Retrieval Prototype System Based-on Ontology,TP391.3
  14. Using Term Synonyms to Extend Bayesian Network Model for Structured Document Retrieval,TP391.3
  15. Based on the similarity of the term extension study Bayesian network retrieval model,TP391.3
  16. Geospatial Data Integration Based on Geo-ontologies,P208
  17. Study and Realization of Optimization of Nutritional Diet to Disease Based on Simulated Annealing Algorithm,R197.324
  18. A Study on the Linguistics Document Categorization Based on Ontology,H0
  19. Traditional Chinese Medical Ontology Based Semantic Relation Discovering and Verification,TP391.1
  20. Research on Folksonomy Based Web Services Discovery,TP393.09

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