A study on reliability in graph discovery

Dai, Honghua 2006, A study on reliability in graph discovery, in Sixth International Conference on Data Mining ICDM 2006 : proceedings : 18-22 December, 2006, Hong Kong, IEEE Computer Society, Los Alamitos, Calif., pp. 775-779.

Attached Files
Name Description MIMEType Size Downloads

Title A study on reliability in graph discovery
Author(s) Dai, Honghua
Conference name International Conference on Data Mining ICDM (6th : 2006 : Hong Kong)
Conference location Hong Kong
Conference dates 18-22 December 2006
Title of proceedings Sixth International Conference on Data Mining ICDM 2006 : proceedings : 18-22 December, 2006, Hong Kong
Editor(s) Clifton, Christopher Wade
Publication date 2006
Conference series International Conference on Data Mining
Start page 775
End page 779
Publisher IEEE Computer Society
Place of publication Los Alamitos, Calif.
Keyword(s) reliable discovery
bayesian network
weak links
small samples
machine learning
reliability
Summary A critical question in data mining is that can we always trust what discovered by a data mining system unconditionally? The answer is obviously not. If not, when can we trust the discovery then? What are the factors that affect the reliability of the discovery? How do they affect the reliability of the discovery? These are some interesting questions to be investigated.

In this paper we will firstly provide a definition and the measurements of reliability, and analyse the factors that affect the reliability. We then examine the impact of model complexity, weak links, varying sample sizes and the ability of different learners to the reliability of graphical model discovery. The experimental results reveal that (1) the larger sample size for the discovery, the higher reliability we will get; (2) the stronger a graph link is, the easier the discovery will be and thus the higher the reliability it can achieve; (3) the complexity of a graph also plays an important role in the discovery. The higher the complexity of a graph is, the more difficult to induce the graph and the lower reliability it would be.
ISBN 0769527027
9780769527024
Language eng
Field of Research 080105 Expert Systems
HERDC Research category E1 Full written paper - refereed
Copyright notice ©2006, IEEE
Persistent URL http://hdl.handle.net/10536/DRO/DU:30006155

Document type: Conference Paper
Collection: School of Engineering and Information Technology
Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Versions
Version Filter Type
Access Statistics: 368 Abstract Views, 5 File Downloads  -  Detailed Statistics
Created: Mon, 07 Jul 2008, 09:59:13 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.