Empirical evaluation of ranking prediction methods for gene expression data classification

Authors:
Bruno Feres De Souza;André C. P. L. F. De Carvalho;Carlos Soares
Affiliations:
ICMC, Universidade de São Paulo, São Carlos, Brazil;ICMC, Universidade de São Paulo, São Carlos, Brazil;FEP, Universidade do Porto, Porto, Portugal
Venue:
IBERAMIA'10 Proceedings of the 12th Ibero-American conference on Advances in artificial intelligence
Year:
2010

Citing 3
Cited 1

Bagging predictors

Machine Learning
Ranking with Predictive Clustering Trees

ECML '02 Proceedings of the 13th European Conference on Machine Learning
Metalearning: Applications to Data Mining

Metalearning: Applications to Data Mining

Analysis of complexity indices for classification problems: Cancer gene expression data

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recently, meta-learning techniques have been employed to the problem of algorithm recommendation for gene expression data classification. Due to their flexibility, the advice provided to the user was in the form of rankings, which are able to express a preference order of Machine Learning algorithms accordingly to their expected relative performance. Thus, choosing how to learn accurate rankings arises as a key research issue. In this work, the authors empirically evaluated 2 general approaches for ranking prediction and extended them. The results obtained for 49 publicly available microarray datasets indicate that the extensions introduced were very beneficial to the quality of the predicted rankings.