Evaluation of Feature Subset Selection, Feature Weighting, and Prototype Selection for Biomedical Applications

Authors:
Suzanne Little;Ovidio Salvetti;Petra Perner
Affiliations:
Institute of Computer Vision and Applied Computer Sciences, Germany;ISTI-CNR, Pisa, Italy;Institute of Computer Vision and Applied Computer Sciences, Germany
Venue:
ECCBR '08 Proceedings of the 9th European conference on Advances in Case-Based Reasoning
Year:
2008

Citing 9
Cited 0

Instance-Based Learning Algorithms

Machine Learning
A comparison between neural networks and decision trees based on data from industrial radiographic testing

Pattern Recognition Letters - Special issue on machine learning and data mining in pattern recognition
Better Prediction of Protein Cellular Localization Sites with the it k Nearest Neighbors Classifier

Proceedings of the 5th International Conference on Intelligent Systems for Molecular Biology
Modelling the Competence of Case-Bases

EWCBR '98 Proceedings of the 4th European Workshop on Advances in Case-Based Reasoning
Weighting Features

ICCBR '95 Proceedings of the First International Conference on Case-Based Reasoning Research and Development
Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)

Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)
Finding Prototypes For Nearest Neighbor Classifiers

IEEE Transactions on Computers
Prototype-based classification

Applied Intelligence
Data mining on multimedia data

Data mining on multimedia data

Quantified Score

Hi-index	0.00

Visualization

Abstract

Many medical diagnosis applications are characterized by datasets that contain under-represented classes due to the fact that the disease appears more rarely than the normal case. In such a situation classifiers that generalize over the data such as decision trees and Naïve Bayesian are not the proper choice as classification methods. Case-based classifiers that can work on the samples seen so far are more appropriate for such a task. We propose to calculate the contingency table and class specific evaluation measures despite the overall accuracy for evaluation purposes of classifiers for these specific data characteristics. We evaluate the different options of our case-based classifier and compare the performance to decision trees and Naïve Bayesian. Finally, we give an outlook for further work.