Download:
|
by P. W. Eklund, A. Hoang
http://www.int.gu.edu.au/kvo/reports/lmdt.ps.gz
Add To MetaCart
Abstract:
This paper records the classification performance of the LMDT algorithm on 26 Irvine datasets. In a shoot-out with other well-known supervised learning algorithms, LMDT is the best classifier in the majority of cases. This work seeks to examine the reasons for these results, in data analysis terms, and uses the visualisation tool, Xgobi, to generate projections of the datasets in support of our explanations. Keywords: LMDT, classifier comparision, visualisation, Xgobi. 1
Citations
|
655
|
UCI Repository of Machine Learning Databases [machine-readable data repository
– Murphy, Aha
- 1992
|
|
619
|
The CN2 Induction Algorithm
– Clark, Niblett
- 1989
|
|
263
|
Rule induction with CN2: Some recent improvements
– CLARK, BOSWELL
- 1991
|
|
86
|
Decision tree induction based on efficient tree restructuring
– Utgoff, Berkman, et al.
- 1997
|
|
74
|
XGobi: Interactive dynamic graphics in the X window system with a link to S
– Swayne, Cook, et al.
- 1991
|
|
34
|
Combining classification algorithms
– Gama
- 1999
|
|
31
|
Multivariate versus univariate decision trees
– Brodley, Utgoff
- 1992
|
|
4
|
Linear machine decision tree (technical report 91-10
– Utgoff, Brodley
- 1991
|
|
3
|
Characterizing the applicability of classification algorithms using meta level learning
– P, Henery
- 1994
|
|
3
|
A system for induction of oblique decision trees
– S
- 1994
|
|
3
|
Xgobi meets S: Integrating software for data analysis
– A
- 1991
|
|
3
|
Manual controls for high-dimensional data projection
– D
- 1996
|
|
3
|
VisDB: Database exploration using multidimentional visualization. Computer Graphic and Application
– P
- 1994
|
|
2
|
A Kolmogorof-Smirnof metric for decision tree induction (technical report 96-3
– Utgoff
- 1996
|
|
2
|
The visual design and control of trellis display
– unknown authors
- 1996
|
