Title :
Accurate Prediction of Transition Energies in Organic Molecules
Author :
Ting Gao ; Hui Li ; Ying-hua Lu ; Hai-Bin Li ; Hong-Zhi Li ; Zhong-Min Su
Author_Institution :
Sch. of Comput. Sci. & Inf. Technol., Northeast Normal Univ., Changchun, China
Abstract :
Least squares support vector machines (LSSVM) has been carried out in order to obtain a statistically meaningful analysis of the extended set of molecules. The combined HF with LSSVM correction approach (LSSVM/HF) has been applied to evaluate the transition energies of organic molecules. After LSSVM correction, the RMS deviations of the calculated transition energies reduce from 0.91 to 0.26 eV for HF methods. And, this LSSVM/HF is a excellent method to predict transition energies and extend the reliably and efficiently of calculated transition energies.
Keywords :
chemistry computing; statistical analysis; support vector machines; HF methods; RMS deviations; least squares support vector machines; organic molecules; statistically meaningful analysis; transition energy prediction; Accuracy; Artificial neural networks; Hafnium; Heating; Support vector machines; Testing; Training; HF; Least squares support vector machines; transition energies;
Conference_Titel :
Frontier of Computer Science and Technology (FCST), 2010 Fifth International Conference on
Conference_Location :
Changchun, Jilin Province
Print_ISBN :
978-1-4244-7779-1
DOI :
10.1109/FCST.2010.9