Pattern recognition system based on support vector machines: HIV-1 integrase inhibitors application

Abstract

Support Vector Machines (SVM) represent one of the most promising Machine Learning (ML) tools that can be applied to develop a predictive Quantitative Structure-Activity Relationship (QSAR) models using molecular descriptors. The performance and predictive power of support vector machines (SVM) for regression problems in quantitative structure-activity relationship were investigated. The SVM results are superior to those obtained by artificial neural network and multiple linear regression. These results indicate that the SVM model with the kernel radial basis function can be used as an alternative tool for regression problems in quantitative structure-activity relationship.

Keywords: Support Vector Machines; Artificial Neural Network; Quantitative Structure-Activity Relationship.