* Invited and Regular Papers will be published in various Indexed Journals (ISI, SCOPUS, EBSCO, DBLP, ACM, IET/INSPEC, Proquest, Copernicus etc...) based on quality and reviewers' recommendations or in our Books in Springer Verlag. Click here
* Authors will be informed about the title of the Journal before
their registration.
* Contact us by email:
support@inase.org Registration fees
Publication Ethics and Malpractice Statement
Deadlines
(Previous Conference: ETE 2016, Vienna, Austria, January 15-17, 2016)
PLENARY SPEAKERS:
Prof. Dimitrios A. Karras, Sterea Hellas Institute of Technology, Dept. Automation, GREECE, e-mail: dakarras2010@gmail.com
Title: "Dealing with Uncertainty in Knowledge Representation"
Abstract: With the rapid growth in the quantity and complexity of scientific knowledge available, the problems associated with dealing with this knowledge are well-recognized. Some of these problems are a result of the uncertainties and inconsistencies that arise in this knowledge. Other problems arise from heterogeneous and informal formats for this knowledge. To address these problems, developments in the application of knowledge representation and reasoning technologies can allow scientific knowledge to be captured in logic-based formalisms. Using such formalisms, we can undertake reasoning with the uncertainty and inconsistency to allow automated techniques to be used for querying and combining of scientific knowledge. Furthermore, by binding background knowledge, the querying and combining tasks can be carried out more judiciously. In this speech, we review some of the significant proposals for formalisms for representing and reasoning with uncertainty, paying special attention to new techniques like interval analysis.
Uncertainty can be classified into four classes: epistemic, linguistic, ambiguity and variability. Epistemic uncertainty is uncertainty due to lack of knowledge. It is also referred to as state of subjective uncertainty, reducible uncertainty, Type B or knowledge uncertainty means uncertainty can be reduced through more relevant data and includes systematic error, subjective uncertainty and measurement uncertainty. The epistemic uncertainty can be represented in many ways including probabilistic theory, fuzzy set, possibility theory, etc. Linguistic uncertainty produced by statements in natural language. Linguistic uncertainty can be classified into five distinct types: vagueness, ambiguity, context dependence, under specificity and indeterminacy of theoretical terms. Ambiguity can be defined as uncertainty about the probability which is created by lack of information that is relevant and could be known. Ambiguity uncertainty is applied to the situation where the probabilities of the outcome are unknown. Variability as name specifies is due to variations or differences in a process or quantity. Assigning an exact value for a quantity is difficult because it depends on many parameters. The variability can be categorized as: natural variation which occurs naturally and inherent randomness are repeated processes and are irregular by nature. Uncertainty analysis is a process that measures, recognizes, identifies and minimizes all types of uncertainty in risk estimates and separates this uncertainty among the risk factors that contributes to relevant risk estimates. The uncertainty analysis includes many statistical problems such as: uncertainty factor, decision making with uncertain information, estimation of uncertainty in complex models of risk, structural uncertainty and model specification and monitoring methods to reduce uncertainty. Quantitative approaches to measure uncertainty vary with complexity of the problem and the associated methods to reduce risk.
This plenary lecture will, therefore, deal with all major tools for handling uncertainty in knowledge representation, including probability theory, fuzzy sets, possibility theory, statistical estimation but special attention will be given to interval analysis as a new and very promising tool for dealing with uncertainty.