Cost-conscious classifier ensembles

Authors:
Cigdem Demir;Ethem Alpaydin
Affiliations:
Department of Computer Engineering, Bogazici University, Istanbul TR-34342, Turkey;Department of Computer Engineering, Bogazici University, Istanbul TR-34342, Turkey
Venue:
Pattern Recognition Letters
Year:
2005

Citing 5
Cited 11

Original Contribution: Stacked generalization

Neural Networks
On Combining Classifiers

IEEE Transactions on Pattern Analysis and Machine Intelligence
Combining Multiple Representations and Classifiers for Pen-based Handwritten Digit Recognitio

ICDAR '97 Proceedings of the 4th International Conference on Document Analysis and Recognition
Adaptive mixtures of local experts

Neural Computation
Cost-sensitive classification: empirical evaluation of a hybrid genetic decision tree induction algorithm

Journal of Artificial Intelligence Research

Using boosting to prune bagging ensembles

Pattern Recognition Letters
Test-Cost Sensitive Classification Based on Conditioned Loss Functions

ECML '07 Proceedings of the 18th European conference on Machine Learning
Using Boosting to prune Double-Bagging ensembles

Computational Statistics & Data Analysis
Incremental construction of classifier and discriminant ensembles

Information Sciences: an International Journal
Cost-conscious multiple kernel learning

Pattern Recognition Letters
Cost-sensitive classification with respect to waiting cost

Knowledge-Based Systems
Neuro-fuzzy-combiner: an effective multiple classifier system

International Journal of Knowledge Engineering and Soft Data Paradigms
Qualitative test-cost sensitive classification

Pattern Recognition Letters
Classifier ensemble construction with rotation forest to improve medical diagnosis performance of machine learning algorithms

Computer Methods and Programs in Biomedicine
Lightweight power aware and scalable movement monitoring for wearable computers: a mining and recognition technique at the fingertip of sensors

Proceedings of the 2nd Conference on Wireless Health
Eigenclassifiers for combining correlated classifiers

Information Sciences: an International Journal

Quantified Score

Hi-index	0.10

Visualization

Abstract

Ensemble methods improve the classification accuracy at the expense of testing complexity, resulting in increased computational costs in real-world applications. Developing a utility-based framework, we construct two novel cost-conscious ensembles; the first one determines a subset of classifiers and the second dynamically selects a single classifier. Both ensembles successfully switch between classifiers according to the accuracy-cost trade-off of an application.