Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts




НазваниеSupport for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts
страница5/35
Дата05.10.2012
Размер0.88 Mb.
ТипДокументы
1   2   3   4   5   6   7   8   9   ...   35

1.4Specific Outcomes


This research has produced the following outcomes.

  • An approach for assessing graphical modelling.

A novel approach and its materialisation are proposed to support the BNKE process through visualisation and verbal explanation of the qualitative part of BNs independently of the quantitative part.

  • Proposed materialisation of the novel approach.
    A specific functionality of the approach and a structured execution of the functionality are proposed as a methodology. This methodology will further reduce the need for an interaction with a BN expert.

  • A prototype tool called V-NET to evaluate the approach and its materialisation.
    This tool is based on the concepts of the materialisation of the approach; in other words, it is built on the practical implementation foundations of the approach, which by itself is abstract.

  • Support for the BN Knowledge-Engineering process

Using V-NET, the proposed materialisation of the approach proved to be useful and satisfactory with regard to its goals:

    • Identification and classification of possible errors in all structural aspects. The tool was used to highlight independencies that might be erroneous, thus helping in the correct construction of a structure/graph by a DE who is not a BN expert.

    • Identification of relationships and new structures, thus improving the understanding of the domain and communicating the BN structure to the DE.

    • Identification of misunderstandings with regard to what relationships the BN represents and how, thus communicating BN technology to the DE.

    • The visualisation and verbal explanation provided by the tool enabled a rapid analysis of a large number of structures. This can be useful in several situations: (i) analysing networks that were produced by machine-learning algorithms. When machine-learning methods are used the output can include many possible structures. (ii) Analysing networks that were derived from ontology. When ontology is used to capture all structural assumptions, several alternative graphical structures can be derived. (iii) Support teaching tasks. When a teacher needs to check networks produced by students as assignments, they have to analyse many structures.

    • The proposed approach offers the advantage of reducing the need for an interaction with a BN expert.



1.5 Overview of Thesis



This thesis is organised as follows:


In chapter 2 I review the current literature, covering the mathematical basis on which BNs are based, and describing the different aspects of constructing a BN, the problems involved, the existing tools to support the process and what is missing.


In chapter 3 I propose an approach for further supporting of the KE process and the materialisation of this approach.


In chapter 4 I describe a tool I constructed (V-NET) to evaluate some aspects of this approach.


In chapter 5 I present and analyse the results of the evaluation of the approach, through V-NET, using qualitative case studies done in different settings. Chapter 5 is concluded with a discussion on the outcome of the evaluation, particularly with regard to the provision of practical answers to the questions posed in section 1.2.


The conclusions from my research and the suggested future work are presented in chapter 6.

2 Literature Review



This thesis describes a proposed approach for investigating the structure of Bayesian networks (BNs) and supporting the knowledge engineering (KE) process. To this end, in this chapter I review of the background literature regarding the various aspects of BNs, associated tools and methods of explaining BN models.


I start with a description of the mathematical background for models in artificial intelligence. In particular, I focus on BNs, models that are based on probability theory and graph theory. I review the process of constructing BNs, i.e. the KE process. I describe both aspects of the KE process, namely knowledge elicitation from experts and learning the structure from data using algorithms, which is a different aspect of the process of KE of BNs.


I then review the existing BN software tools, describing their advantages and limitations. Finally, I conclude with a description of explanation methods in BNs, since they can be utilised in the KE process of BNs.

1   2   3   4   5   6   7   8   9   ...   35

Похожие:

Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconThis document is being distributed by the American International Health Alliance (aiha) / the EurasiaHealth Knowledge Network with support from the us agency
Саратовский областной Центр по профилактике и борьбе со спид и инфекционными заболеваниями
Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconStudents will demonstrate competence in; Engineering Curriculum; basic concepts such as significant digits, length, and mass; Engineering Drawing; the use of the computer as an engineering tool; designing a project

Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconAn Object-Oriented Support Tool for the Design of Casting Procedures

Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconRegional network of co-operation in Europe as a tool of efficient governance

Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconEnabling Decision Support and Costing of Product Designs by using Visual Metaphors

Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconRepresenting and using legal knowledge in integrated decision support systems: DataLex WorkStations

Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconThis document contains a list of references to publications and reports about Bayesian Net technology, and especially Bayesian Net applications. The report will

Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconNorwegian Network for the Support of the Indigenous Peoples of the Russian Arctic (nnsipra) Сеть Норвежских Организаций в Поддержку Коренных Народов Российского Севера

Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconDevelopment of a Product model to support engineering change management

Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts iconTool Use: Misuse of shop tools will result in the loss of tool privileges

Разместите кнопку на своём сайте:
Библиотека


База данных защищена авторским правом ©lib.znate.ru 2014
обратиться к администрации
Библиотека
Главная страница