miun.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Methods for Evaluating Multi-Level Decision Trees in Imprecise Domains
Mid Sweden University, Faculty of Science, Technology and Media, Department of Information Technology and Media.
Responsible organisation
2005 (English)In: Proceedings of the Eighteenth International Florida Artificial Intelligence Research Society Conference, FLAIRS 2005 - Recent Advances in Artifical Intelligence, AAAI Press, 2005, 734-739 p.Conference paper, (Other academic)
Abstract [en]

Over the years numerous decision analytical models based on interval estimates of probabilities and utilities have been developed, and a few of these models have also been implemented into software. However, only one software, the Delta approach, are capable of handling probabilities, values and weights simultaneously, and also allow for omparative relations, which are very useful where the information quantity is limited. A major disadvantage with this method is that it only allows for single-level decision trees and cannot non-trivially be extended to handle multi-level decision trees. This paper generalizes the Delta approach into a method for handling multi-level decision trees. The straight-forward way of doing this is by using a multi-linear solver; however, this is very demanding from a computational point of view. The proposed solutions are instead to either recursively collapse the multi-level decision tree into a single-level tree or, preferably, use backward induction, thus mapping it to a bilinear problem. This can be solved by LP-based algorithms, which facilitate reasonable computational effort.

Place, publisher, year, edition, pages
AAAI Press, 2005. 734-739 p.
National Category
Computer Science
Identifiers
URN: urn:nbn:se:miun:diva-3782Scopus ID: 2-s2.0-32844468538Local ID: 4074ISBN: 1577352343 (print)OAI: oai:DiVA.org:miun-3782DiVA: diva2:28814
Conference
Recent Advances in Artifical Intelligence - Eighteenth International Florida Artificial Intelligence Research Society Conference, FLAIRS 2005; Clearwater Beach, FL; United States; 15 May 2005 through 17 May 2005; Code 66624
Available from: 2008-11-22 Created: 2008-11-22 Last updated: 2016-09-27Bibliographically approved
In thesis
1. An applied approach to numerically imprecise decision making
Open this publication in new window or tab >>An applied approach to numerically imprecise decision making
2007 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Despite the fact that unguided decision making might lead to inefficient and nonoptimal decisions, decisions made at organizational levels seldom utilise decisionanalytical tools. Several gaps between the decision-makers and the computer baseddecision tools exist, and a main problem in managerial decision-making involves the lack of information and precise objective data, i.e. uncertainty and imprecision may be inherent in the decision situation. We believe that this problem might be overcome by providing computer based decision tools capable of handling the uncertainty inherent in real-life decision-making. At present, nearly all decision analytic software is only able to handle precise input, and no known software is capable of handling full scale imprecision, i.e. imprecise probabilities, values and weights, in the form of interval and comparative statements. There are, however, some theories which are able to handle some kind of uncertainty, and which deal with computational and implementational issues, but if they are never actually operationalised, they are of little real use for a decision-maker. Therefore, a natural question is how a reasonable decision analytical framework can be built based on prevailing interval methods, thus dealing with the problems of uncertain and imprecise input? Further, will the interval approach actually prove useful? The framework presented herein handles theoretical foundations for, and implementations of, imprecise multi-level trees, multi-criteria, risk analysis, together with several different evaluation options. The framework supports interval probabilities, values, and criteria weights, as well as comparative statements, also allowing for mixing probabilistic and multi-criteria decisions. The framework has also been field tested in a number of studies, proving the usefulness of the interval approach.

Place, publisher, year, edition, pages
Sundsvall: Mid Sweden University, 2007. 50 p.
Series
Mid Sweden University doctoral thesis, ISSN 1652-893X ; 21
Keyword
decision tree, decision analysis, decision tool, imprecise reasoning, probability
National Category
Computer Science
Identifiers
urn:nbn:se:miun:diva-7147 (URN)978-91-85317-49-3 (ISBN)
Public defence
(English)
Opponent
Supervisors
Available from: 2008-11-29 Created: 2008-11-23 Last updated: 2009-02-13Bibliographically approved

Open Access in DiVA

No full text

Scopus

Search in DiVA

By author/editor
Johansson [Idefeldt], Jim
By organisation
Department of Information Technology and Media
Computer Science

Search outside of DiVA

GoogleGoogle Scholar

Total: 217 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf