Abstract
High-throughput biological technologies offer the promise of finding feature sets to serve as biomarkers for medical applications; however, the sheer number of potential features (genes, proteins, etc.) means that there needs to be massive feature selection, far greater than that envisioned in the classical literature. This paper considers performance analysis for feature-selection algorithms from two fundamental perspectives: How does the classification accuracy achieved with a selected feature set compare to the accuracy when the best feature set is used and what is the optimal number of features that should be used? The criteria manifest themselves in several issues that need to be considered when examining the efficacy of a feature-selection algorithm: (1) the correlation between the classifier errors for the selected feature set and the theoretically best feature set; (2) the regressions of the aforementioned errors upon one another; (3) the peaking phenomenon, that is, the effect of sample size on feature selection; and (4) the analysis of feature selection in the framework of high-dimensional models corresponding to high-throughput data.
Current Genomics
Title: Performance of Feature Selection Methods
Volume: 10 Issue: 6
Author(s): Edward R. Dougherty, Jianping Hua and Chao Sima
Affiliation:
Abstract: High-throughput biological technologies offer the promise of finding feature sets to serve as biomarkers for medical applications; however, the sheer number of potential features (genes, proteins, etc.) means that there needs to be massive feature selection, far greater than that envisioned in the classical literature. This paper considers performance analysis for feature-selection algorithms from two fundamental perspectives: How does the classification accuracy achieved with a selected feature set compare to the accuracy when the best feature set is used and what is the optimal number of features that should be used? The criteria manifest themselves in several issues that need to be considered when examining the efficacy of a feature-selection algorithm: (1) the correlation between the classifier errors for the selected feature set and the theoretically best feature set; (2) the regressions of the aforementioned errors upon one another; (3) the peaking phenomenon, that is, the effect of sample size on feature selection; and (4) the analysis of feature selection in the framework of high-dimensional models corresponding to high-throughput data.
Export Options
About this article
Cite this article as:
Dougherty R. Edward, Hua Jianping and Sima Chao, Performance of Feature Selection Methods, Current Genomics 2009; 10 (6) . https://dx.doi.org/10.2174/138920209789177629
DOI https://dx.doi.org/10.2174/138920209789177629 |
Print ISSN 1389-2029 |
Publisher Name Bentham Science Publisher |
Online ISSN 1875-5488 |
Call for Papers in Thematic Issues
Advanced Computational Algorithms and Artificial Intelligence in Clinical Pharmacogenomics
In the era of personalized medicine, understanding the relationship between genetics and drug response is crucial. This issue delves into innovative methodologies, leveraging deep computational analysis and artificial intelligence, to enhance the field of Clinical Pharmacogenomics. The interdisciplinary approach harnesses the power of advanced high-throughput genotyping technologies, sophisticated computational analysis, ...read more
Applications of Single-cell Sequencing Technology in Reproductive Medicine
Single cell sequencing (SCS) technology utilizes individual cells' genetic material to sequence their genome, transcriptome, and epigenetics at the molecular level. It offers insights into cell heterogeneity and enables the study of limited biological materials. Since its recognition as a valuable technique in 2011, single cell sequencing has yielded numerous ...read more
Big Data in Cancer Research
Cancer is a significant threat to human life and health, remaining a highly aggressive killer. It is a leading cause of death worldwide and represents a crucial medical issue for humanity. However, in the past decade, the effectiveness of new synthetic anticancer agents has not matched the current clinical speculation. ...read more
Current Genomics in Cardiovascular Research
Cardiovascular diseases are the main cause of death in the world, in recent years we have had important advances in the interaction between cardiovascular disease and genomics. In this Research Topic, we intend for researchers to present their results with a focus on basic, translational and clinical investigations associated with ...read more
Related Journals
- Author Guidelines
- Graphical Abstracts
- Fabricating and Stating False Information
- Research Misconduct
- Post Publication Discussions and Corrections
- Publishing Ethics and Rectitude
- Increase Visibility of Your Article
- Archiving Policies
- Peer Review Workflow
- Order Your Article Before Print
- Promote Your Article
- Manuscript Transfer Facility
- Editorial Policies
- Allegations from Whistleblowers
- Announcements
Related Articles
-
Advances in Cancer Stem Cell Therapy: Targets and Treatments
Recent Patents on Regenerative Medicine Advances in Helper-Dependent Adenoviral Vector Research
Current Gene Therapy Antisense Oligonucleotides as an Innovative Therapeutic Strategy in the Treatment of High-Grade Gliomas
Recent Patents on CNS Drug Discovery (Discontinued) Intracellular Delivery of Potential Therapeutic Genes: Prospects in Cancer Gene Therapy
Current Gene Therapy Leptomeningeal Metastasis: Challenges in Diagnosis and Treatment
Current Cancer Therapy Reviews Strategies of overcoming the physiological barriers for tumor-targeted nano-sized drug delivery systems
Current Pharmaceutical Design Lipid Based Anti-Retroviral Nanocarriers: A Review of Current Literature and Ongoing Studies
Drug Delivery Letters Physiological and Pathological Functions of Acid-Sensing Ion Channels in the Central Nervous System
Current Drug Targets Biological Activities of Eco-Friendly Synthesized Hantzsch Adducts
Medicinal Chemistry Smac-Derived Aza-Peptide As an Aminopeptidase-Resistant XIAP BIR3 Antagonist
Protein & Peptide Letters Safety and Utilization of Blood Components as Therapeutic Delivery Systems
Current Pharmaceutical Biotechnology Apoptotic Signaling in Pancreatic Cancer – Therapeutic Application (Supplemental Data)
Current Cancer Therapy Reviews Cytotoxic Activity of Polysubstituted 7-chloro-4-quinolinylhydrazone Derivatives
Letters in Drug Design & Discovery A Comparison of Physicochemical Property Profiles of Marketed Oral Drugs and Orally Bioavailable Anti-Cancer Protein Kinase Inhibitors in Clinical Development
Current Topics in Medicinal Chemistry Imidazoles and Benzimidazoles as Tubulin-Modulators for Anti-Cancer Therapy
Current Medicinal Chemistry Recent Innovations in Antibody-Mediated, Targeted Particulate Nanotechnology and Implications for Advanced Visualisation and Drug Delivery
Current Nanoscience Biomedical Technologies for In Vitro Screening and Controlled Delivery of Neuroactive Compounds
Central Nervous System Agents in Medicinal Chemistry Current Understanding of Epigenetics Driven Therapeutic Strategies in Colorectal Cancer Management
Endocrine, Metabolic & Immune Disorders - Drug Targets Advances in Molecular Therapeutic Approaches to Patients with Malignant Gliomas
Current Signal Transduction Therapy Anticancer Agent Ukrain and Bortezomib Combination is Synergistic in 4T1 Breast Cancer Cells
Anti-Cancer Agents in Medicinal Chemistry