Article

On biases in estimating multi-valued attributes

Author:

IJCAI'95: Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2

Pages 1034 - 1040

Published: 20 August 1995 Publication History

Abstract

We analyse the basics of eleven measures for estimating the quality of the multivalued attributes. The values of information gain J-measure, gini-index and relevance tend to lin early increase with the number of values of an attribute. The values of gam-ratio dis tance measure, Relief and the weight of evidence decrease for informative attributes and increase for irrelevant attributes. The bias of the statistic tests based on the chi-square distribution is similar but these functions are not able to discriminate among the attributes of different quality. We also introduce a new func tion based on the MDL principle whose value slightly decreases with the increasing number of attributes values.

References

[1]

[Bairn 1988] PW Bairn A method for attribute selection in inductive learning systems, IEEE Trans on PAMI, 10 888-896, 1988.

Digital Library

Google Scholar

[2]

[Breiman et al 1984] L Breiman J H Friedman R A Olshen and C J Stone Classification and Regression Trees Wadsworth International Group 1984.

Google Scholar

[3]

[Kira and Rendell, 1992] K Kira and L Rendell A practical approach to feature selection Proc Intern Conf on Machine Learning (Aberdeen, July 1992). D Sleeman and P Edwards (eds), Morgan Kaufmann pp 249-256.

Crossref

Google Scholar

[4]

[Kononenko 1994] I Kononenko Estimating attributes Analysis and extensions of RELIEF Proc European Conf on Machine Learning (Catania April 1994) L Dr Raedt and F Bergadano (eds) Springer Verlag pp 171-182.

Crossref

Google Scholar

[5]

[Li and Yilanyi 1993] M Li and P Yilanyi An introduction to holmogoroi Complexity and its applications, Springer Verlag, 1993.

Crossref

Google Scholar

[6]

[Mantaras 1989] RL Mantaras ID3 Revisited A distance based criterion for attribute selection Proc Int Symp Methodologies for Intelligent Systems, Char lotte, North Carolina USA, Oct 1989.

Google Scholar

[7]

[Michie 1989] D Michie Personal Models of Rationality J of Statistical Planning and Inference, Special Issue on Foundations and Philosophy of Probability and Statistics 21 1989.

Google Scholar

[8]

[Press et al, 1988] W H Press, S A Teukolsky W T Vetterling B P Flannery Numerical recipes in C The art of scientific computing, Cambridge University Press, 1988.

Digital Library

Google Scholar

[9]

[Quinlan 1986] R Quinlan Induction of decision trees Machine Learning, 1 81-106, 1986.

Crossref

Google Scholar

[10]

[Rissanen, 1983] J R Rissanen J A universal prior for integers and estimation by minimum description length Annals of Statistics, 11 416-431, 1983.

Google Scholar

[11]

[Smyth and Goodman, 1990] P Smyth and R M Goodman Rule induction using information theory. In G Piatetsky Shapiro and W Frawley (eds) Kmowledge Discovery in Databases, MIT Press, 1990.

Google Scholar

[12]

[White and Liu, 1994] A P White and W Z Liu Bias in information-based measures in decision tree induction Machine Learning 15 121-329, 1994.

Crossref

Google Scholar

Cited By

View all

Jimenez MRwemalika RPapadakis MSarro FLe Traon YHarman MDumas MPfahl DApel SRusso A(2019)The importance of accounting for real-world labelling when predicting software vulnerabilitiesProceedings of the 2019 27th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering10.1145/3338906.3338941(695-705)Online publication date: 12-Aug-2019
https://dl.acm.org/doi/10.1145/3338906.3338941
Papamartzivanos DGmez Mrmol FKambourakis G(2018)DendronFuture Generation Computer Systems10.1016/j.future.2017.09.05679:P2(558-574)Online publication date: 1-Feb-2018
https://dl.acm.org/doi/10.1016/j.future.2017.09.056
Baron G(2016)On Influence of Representations of Discretized Data on Performance of a Decision SystemProcedia Computer Science10.1016/j.procs.2016.08.18796:C(1418-1427)Online publication date: 1-Oct-2016
https://dl.acm.org/doi/10.1016/j.procs.2016.08.187
Show More Cited By

Recommendations

Bias of importance measures for multi-valued attributes and solutions
ICANN'11: Proceedings of the 21st international conference on Artificial neural networks - Volume Part II

Attribute importance measures for supervised learning are important for improving both learning accuracy and interpretability. However, it is well-known there could be bias when the predictor attributes have different numbers of values. We propose two ...
Towards estimating the number of distinct value combinations for a set of attributes
CIKM '05: Proceedings of the 14th ACM international conference on Information and knowledge management

Accurately and efficiently estimating the number of distinct values for some attribute(s) or sets of attributes in a data set is of critical importance to many database operations, such as query optimization and approximation query answering. Previous ...
Estimating attributes: Analysis and extensions of RELIEF
ECML'94: Proceedings of the 7th European Conference on Machine Learning

In the context of machine learning from examples this paper deals with the problem of estimating the quality of attributes with and without dependencies among them. Kira and Rendell (1992a,b) developed an algorithm called RELIEF, which was shown to be ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

IJCAI'95: Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2

August 1995

2077 pages

ISBN:1558603638

Publisher

Morgan Kaufmann Publishers Inc.

San Francisco, CA, United States

Publication History

Published: 20 August 1995

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

30
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 12 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

View all

Jimenez MRwemalika RPapadakis MSarro FLe Traon YHarman MDumas MPfahl DApel SRusso A(2019)The importance of accounting for real-world labelling when predicting software vulnerabilitiesProceedings of the 2019 27th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering10.1145/3338906.3338941(695-705)Online publication date: 12-Aug-2019
https://dl.acm.org/doi/10.1145/3338906.3338941
Papamartzivanos DGmez Mrmol FKambourakis G(2018)DendronFuture Generation Computer Systems10.1016/j.future.2017.09.05679:P2(558-574)Online publication date: 1-Feb-2018
https://dl.acm.org/doi/10.1016/j.future.2017.09.056
Baron G(2016)On Influence of Representations of Discretized Data on Performance of a Decision SystemProcedia Computer Science10.1016/j.procs.2016.08.18796:C(1418-1427)Online publication date: 1-Oct-2016
https://dl.acm.org/doi/10.1016/j.procs.2016.08.187
Almuhaideb SMenai M(2016)Impact of preprocessing on medical data classificationFrontiers of Computer Science: Selected Publications from Chinese Universities10.1007/s11704-016-5203-510:6(1082-1102)Online publication date: 1-Dec-2016
https://dl.acm.org/doi/10.1007/s11704-016-5203-5
Ferreira AFigueiredo M(2014)Incremental filter and wrapper approaches for feature discretizationNeurocomputing10.1016/j.neucom.2012.10.036123(60-74)Online publication date: 1-Jan-2014
https://dl.acm.org/doi/10.1016/j.neucom.2012.10.036
Fernandes JLozano JInza IIrigoien XPérez ARodríguez J(2013)Supervised pre-processing approaches in multiple class variables classification for fish recruitment forecastingEnvironmental Modelling & Software10.5555/2773797.277387940:C(245-254)Online publication date: 1-Feb-2013
https://dl.acm.org/doi/10.5555/2773797.2773879
Musliu NSchwengerer M(2013)Algorithm Selection for the Graph Coloring ProblemRevised Selected Papers of the 7th International Conference on Learning and Intelligent Optimization - Volume 799710.1007/978-3-642-44973-4_42(389-403)Online publication date: 7-Jan-2013
https://dl.acm.org/doi/10.1007/978-3-642-44973-4_42
Lakoumentas JDrakos JKarakantza MSakellaropoulos GMegalooikonomou VNikiforidis G(2012)Optimizations of the naïve-Bayes classifier for the prognosis of B-Chronic Lymphocytic Leukemia incorporating flow cytometry dataComputer Methods and Programs in Biomedicine10.1016/j.cmpb.2012.02.009108:1(158-167)Online publication date: 1-Oct-2012
https://dl.acm.org/doi/10.1016/j.cmpb.2012.02.009
Wu TXie KNie CSong G(2012)An adaption of relief for redundant feature eliminationProceedings of the 9th international conference on Advances in Neural Networks - Volume Part II10.1007/978-3-642-31362-2_9(73-81)Online publication date: 11-Jul-2012
https://dl.acm.org/doi/10.1007/978-3-642-31362-2_9
Deng HRunger GTuv E(2011)Bias of importance measures for multi-valued attributes and solutionsProceedings of the 21st international conference on Artificial neural networks - Volume Part II10.5555/2029604.2029642(293-300)Online publication date: 14-Jun-2011
https://dl.acm.org/doi/10.5555/2029604.2029642
Show More Cited By

View Options

View options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Cited By

Recommendations

Bias of importance measures for multi-valued attributes and solutions

Towards estimating the number of distinct value combinations for a set of attributes

Estimating attributes: Analysis and extensions of RELIEF