Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

A new descriptor selection scheme for SVM in unbalanced class problem: a case study using skin sensitisation dataset.

S Li¹, A Fedorowicz, M E Andrew

¹Health Effects Laboratory Division, National Institute for Occupational Safety and Health, Morgantown, WV 26505, USA. sw14@cdc.gov

SAR and QSAR in Environmental Research

|July 27, 2007

Summary

A new method improves Support Vector Machine (SVM) classification for skin sensitization prediction. This descriptor selection scheme achieved 84.9% accuracy, outperforming traditional methods like linear discriminant analysis (LDA).

Related Concept Videos

Skin Cancer

Skin Cancer

Skin cancer is a type of cancer that occurs when there is an abnormal growth of skin cells, usually triggered by damage to the DNA within the skin cells. It is primarily caused by exposure to ultraviolet (UV) radiation from the sun or artificial sources like tanning beds. Skin cancer is the most common type of cancer worldwide, and its incidence continues to rise.
Basal Cell Carcinoma (BCC): BCC is the most common type of skin cancer, accounting for about 80% of cases. It typically develops in...

Classification of Systems-I

Classification of Systems-I

Linearity is a system property characterized by a direct input-output relationship, combining homogeneity and additivity.
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Associations of burnout with awakening and diurnal cortisol among police officers.

Comprehensive psychoneuroendocrinology·2021

Same author

Alterations in NO- and PGI<sub>2</sub>- dependent function in aorta in the orthotopic murine model of metastatic 4T1 breast cancer: relationship with pulmonary endothelial dysfunction and systemic inflammation.

BMC cancer·2018

Same author

Importance of dispersal routes that minimize open-ocean movement to the genetic structure of island populations.

Conservation biology : the journal of the Society for Conservation Biology·2015

Same author

Raman microimaging of murine lungs: insight into the vitamin A content.

The Analyst·2014

Same author

Identification of phenolic dermal sensitizers in a wound closure tape.

Journal of immunotoxicology·2008

Same author

Dimensional analysis of actigraphic derived sleep data.

Nonlinear dynamics, psychology, and life sciences·2008

Area of Science:

Computational chemistry
Toxicology
Machine learning

Background:

Predicting skin sensitization is crucial for chemical safety assessment.
Existing methods for descriptor selection in classification models can be improved.
Support Vector Machines (SVMs) are powerful tools for classification tasks.

Purpose of the Study:

To develop and evaluate a novel descriptor selection scheme for SVM classification.
To optimize SVM performance for skin sensitization prediction using the proposed scheme.
To compare the performance of SVM with different kernels and descriptor selection filters against Fisher's linear discriminant analysis (LDA).

Main Methods:

A backward elimination procedure guided by leave-one-out cross-validation mean accuracy was developed for SVM.

Related Experiment Videos

Descriptor subsets were pre-selected using either a sequential t-test filter or a Random Forest filter.

The performance of different SVM kernels (Radial Basis Function and linear) was evaluated with the descriptor selection scheme.

Fisher's linear discriminant analysis (LDA) was used as a benchmark for comparison.

Main Results:

The proposed descriptor selection scheme, combined with a Radial Basis Function (RBF) kernel and a sequential t-test filter, achieved the highest mean accuracy of 84.9%.
This optimal SVM model utilized 23 descriptors, yielding a sensitivity of 93.1% and a specificity of 76.6%.
A linear kernel with a Random Forest filter provided comparable performance using 24 descriptors.
SVM demonstrated superior performance over LDA when subjected to the same descriptor selection process.

Conclusions:

The novel descriptor selection scheme effectively enhances SVM classification performance for skin sensitization prediction.
SVM, particularly with an RBF kernel and sequential t-test filtering, offers a robust and accurate approach for predicting skin sensitization.
The developed method outperforms traditional LDA, highlighting its potential utility in toxicological assessments.