Download title - VideoLectures.NET

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Computational fluid dynamics wikipedia , lookup

Corecursion wikipedia , lookup

Financial economics wikipedia , lookup

Sociocracy wikipedia , lookup

Data analysis wikipedia , lookup

Pattern recognition wikipedia , lookup

Neuroinformatics wikipedia , lookup

Predictive analytics wikipedia , lookup

Data assimilation wikipedia , lookup

Multidimensional empirical mode decomposition wikipedia , lookup

Multiple-criteria decision analysis wikipedia , lookup

Transcript
Data Mining and Decision Support Integration
Marko Bohanec
Jožef Stefan Institute, Jamova 39, SI-1000 Ljubljana, Slovenia, and
University of Ljubljana, Faculty of Administration, Gosarjeva 5
[email protected]
Abstract
The aim of this presentation is twofold: (1) to introduce the field of Decision
Support (DS), and (2) to provide an overview of possible approaches and
benefits of combining DS with Data Mining (DM) in solving real-life
decision and data-analysis problems. Related to DS, we define the concepts
of decision problem and decision-making, introduce the taxonomy of
disciplines related to DS, overview the approach of decision analysis,
introduce the method of multi-attribute modeling, and illustrate it through
real-life examples of housing loan allocation and risk assessment in
medicine. In the main part, we investigate the ways to combine and integrate
DS and DM, which generally involve the following categories: (1) DS for
DM, (2) DM for DS, (3) DM, then DS, (4) DS, then DM, and (5) DM and
DS. Each category is illustrated by a practical example. Two categories are
investigated in greater detail. The category “(1) DS for DM” is represented
by a method for selecting a best DM-induced classifier based on ROC space
exploration. For the category “(5) DM and DS”, we explore an approach of
developing qualitative multi-attribute models by combining the systems
DEX and HINT. DEX is a DS tool for expert-based (“hand-crafted”)
development of models, whereas HINT is a DM tool that develops models
from data by a method based on function decomposition.
References
BOHANEC, M.: DEXi: A Program for Multi-Attribute Decision Making.
http://www-ai.ijs.si/MarkoBohanec/dexi.html
BOHANEC, M., ZUPAN, B. and RAJKOVIČ, V. (2000): Applications of
qualitative multi-attribute decision models in health care, International
Journal of Medical Informatics 58-59, 191–205.
BOHANEC, M. and ZUPAN, B. (2004): A function-decomposition method
for development of hierarchical multi-attribute decision models. Decision
Support Systems 36, 215–233.
MLADENIČ, D., LAVRAČ, N., BOHANEC, M. and MOYLE, S. (eds.)
(2003): Data Mining and Decision Support: Integration and
Collaboration, KluwerAcademic Publishers, Boston; Dordrecht; London
(Chapters 3, 4, 7, 15, 16, and 17)
MLADENIĆ, D. and LAVRAČ, N. (eds.) (2003): Data Mining and Decision
Support for Business Competitiveness: A European Virtual Enterprise Results of the Sol-Eu-Net Project, DZS, Ljubljana.
ZUPAN, B.: Decisions at Hand. http://www.ailab.si/app/palm/
ZUPAN, B.: Function Decomposition. http://magix.fri.uni-lj.si/hint/
Keywords
Decision support, decision analysis, decision modeling, multi-attribute
decision models, data mining, ROC space, DEX, HINT, concept learning,
function decomposition