Support Vector Machine Classification of Physical and Biological Datasets

Abstract

The support vector machine (SVM) is used in the classification of sonar signals and DNA-binding proteins. Our study on the classification of sonar signals shows that SVM produces a result better than that obtained from other classification methods, which is consistent from the findings of other studies. The testing accuracy of classification is 95.19% as compared with that of 90.4% from multilayered neural network and that of 82.7% from nearest neighbor classifier. From our results on the classification of DNA-binding proteins, one finds that SVM gives a testing accuracy of 82.32%, which is slightly better than that obtained from an earlier study of SVM classification of protein–protein interactions. Hence, our study indicates the usefulness of SVM in the identification of DNA-binding proteins. Further improvements in SVM algorithm and parameters are suggested.

Optimization of process parameters on grain size of Fe<inf>3</inf>O<inf>4</inf> nanoparticles by support vector regression

“…Theory of SVR Support vector machine (SVM) is a powerful tool for data analysis. When SVM is employed to tackle the problems of function approximation and regression estimation by the introduction of an alternative loss function, it is called SVR [12][13][14][15][16][17]. In SVR, the basic idea is to map the x into a higherdimensional feature space F via a nonlinear mapping Φ(x) and then to do a linear regression in this space.…”

Prediction of the Glass Transition Temperatures of Styrenic Copolymers by Using Support Vector Regression Combined with Particle Swarm Optimization

“…At present, it has become a focus in machine learning research and is extensively employed in a wide range of real-word problems. [23][24][25][26][27][28][29][30][31] When SVM is applied to regression by introduction of an alternative loss function, it is termed as SVR. Suppose a sample is described by (x, y), where x represents the independent variable and y the dependent response.…”

SVR-based analysis on tribological property of ultra high molecular weight polyethylene composites filled with nano-ZnO particles

“…It has been widely and successfully used for classification and regression in many real problems [11][12][13][14][15]. SVM is termed as SVR when it was employed for regression.…”