Critical Properties and Acentric Factors of Pure Compounds Modelling Based on QSPR-SVM with Dragonfly Algorithm

This work aimed to model the critical pressure, temperature, volume properties, and acentric factors of 6700 pure compounds based on five relevant descriptors two thermodynamic properties. To that end, four methods were used, namely, multi-linear regression (MLR), artificial neural networks (ANNs), support vector machines (SVMs) using sequential minimal optimisation (SMO), hybrid SVM with Dragonfly algorithm (SVM-DA) each property. The results suggested SVM-DA had better prediction performance compared other models in terms average absolute relative deviation (AARD%) {0.7551, 1.962, 1.929, 2.173} R 2 {0.9699, 0.9673, 0.9856, 0.9766} for volume, factor, respectively. developed can be used estimate property newly designed only from their molecular structure.