Study Of E-Smooth Support Vector Regression And Comparison With E- Support Vector Regression And Potential Support Vector Machines For Prediction For The Antitubercular Activity Of Oxazolines And Oxazoles Derivatives

摘要

A new smoothing method for solving ? -support vector regression (?-SVR),tolerating a small error in fitting a given data sets nonlinearly is proposedin this study. Which is a smooth unconstrained optimization reformulation ofthe traditional linear programming associated with a ?-insensitive supportvector regression. We term this redeveloped problem as ?-smooth support vectorregression (?-SSVR). The performance and predictive ability of ?-SSVR areinvestigated and compared with other methods such as LIBSVM (?-SVR) and P-SVMmethods. In the present study, two Oxazolines and Oxazoles molecular descriptordata sets were evaluated. We demonstrate the merits of our algorithm in aseries of experiments. Primary experimental results illustrate that ourproposed approach improves the regression performance and the learningefficiency. In both studied cases, the predictive ability of the ?- SSVR modelis comparable or superior to those obtained by LIBSVM and P-SVM. The resultsindicate that ?-SSVR can be used as an alternative powerful modeling method forregression studies. The experimental results show that the presented algorithm?-SSVR, plays better precisely and effectively than LIBSVMand P-SVM inpredicting antitubercular activity.