Proceedings Article10.1109/ICIP.2006.312667
Feature Selection using a Mixed-Norm Penalty Function
Huiwen Zeng,H.J. Trussell +1 more
- 01 Oct 2006
- pp 997-1000
4
TL;DR: In this paper, a penalty function combined with a neural network is proposed to select a subset from a collection of features while maintaining the performance possible with the larger set, which is shown to work on test cases with known redundancy.
read more
Abstract: Feature selection is the process of selecting effective subsets of features that are effective in performing a given task. We propose an approach using a penalty function combined with a neural network to select a subset from a collection of features while maintaining the performance possible with the larger set. The penalty function is related to a mixed-norm function that has proven successful in pruning neural networks. The new function is shown to work on test cases with known redundancy and to be effective in feature selection for practical problems.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Comparison of artificial neural network with logistic regression as classification models for variable selection for prediction of breast cancer patient outcomes
V. Bourdes,Stéphane Bonnevay,Paolo Lisboa,Rémy Defrance,David Pérol,Sylvie Chabaud,Thomas Bachelot,Thérèse Gargi,Sylvie Negrier +8 more
TL;DR: The results enhanced the relevance of the use of NN models in predictive analysis in oncology, which appeared to be more accurate in prediction in this French breast cancer cohort.
Constrained Dimensionality Reduction Using a Mixed-Norm Penalty Function with Neural Networks
Huiwen Zeng,H.J. Trussell +1 more
TL;DR: This work shows how to eliminate neurons on that layer of a network and simplify the problem, and introduces a novel penalty function and combine it with bilevel optimization to solve the constrained problem.
12
Multi-metaheuristic Features Selection Model for High-Dimensional Biomedical Data
Saba M. Hussain
- 01 Jan 2021
TL;DR: In this article, a multi-metaheuristic features selection model (MHFS) was proposed to improve the feature selection problems in gene expression profiles. But the proposed system is used the two algorithms to search in parallel for the best number of features that are particle swarm optimization and bat algorithm.
1
References
•Journal Article
An extensive empirical study of feature selection metrics for text classification
TL;DR: An empirical comparison of twelve feature selection methods evaluated on a benchmark of 229 text classification problem instances, revealing that a new feature selection metric, called 'Bi-Normal Separation' (BNS), outperformed the others by a substantial margin in most situations and was the top single choice for all goals except precision.
Multisurface Method of Pattern Separation for Medical Diagnosis Applied to Breast Cytology
TL;DR: The diagnosis of breast cytology is used to demonstrate the applicability ofMultisurface pattern separation to medical diagnosis and decision making and it is found that this mathematical method is applicable to other medical diagnostic and decision-making problems.
1.1K
An efficient fuzzy classifier with feature selection based on fuzzy entropy
Hahn-Ming Lee,Chih-Ming Chen,Jyh-Ming Chen,Yu-Lu Jou +3 more
- 01 Jun 2001
TL;DR: This paper presents an efficient fuzzy classifier with the ability of feature selection based on a fuzzy entropy measure and investigates the use of fuzzy entropy to select relevant features.
319
Algorithms for bilevel optimization
Natalia Alexandrov,John E. Dennis +1 more
- 01 Sep 1994
TL;DR: This work considers necessary optimality conditions for the bilevel problem formulations and discusses results that can be extended to obtain multilevel optimization formulations with constraints at each level.
Dimensionality reduction in hyperspectral image classification
Huiwen Zeng,H.J. Trussell +1 more
- 24 Oct 2004
TL;DR: A novel penalty function is presented and shown to successfully reduce the number of active neurons, which corresponds to the dimensionality of the data for the task of interest.
15