DocumentCode :
3207089
Title :
Improving the Accuracy of Low Level Density Functional Theory Calculation for Absorption Energies: The Least Squares Support Vector Machine
Author :
Gao, Ting ; Pu, Dong-Bing ; Li, Hui ; Lu, Ying-Hua ; Li, Hai-Bin ; Li, Hong-Zhi ; Su, Zhong-Min
Author_Institution :
Sch. of Comput. Sci. & Inf. Technol., Northeast Normal Univ., Changchun, China
Volume :
2
fYear :
2010
fDate :
11-12 May 2010
Firstpage :
522
Lastpage :
525
Abstract :
Least squares support vector machines (LS-SVM) was introduced to improve the calculation accuracy of low level density functional theory. As a demonstration, this combined low level quantum mechanical calculation with LS-SVM correction approach has been applied to evaluate the absorption energies of 160 organic molecules. After LS-SVM correction, the RMS deviations of the calculated absorption energies reduce from 0.95 to 0.16 eV for B3LYP/STO-3G methods. And, this LS-SVM/B3LYP/STO-3G is a excellent method to predict absorption energies and can be used as an alternative to experimental measurements and high-accurate quantum chemical calculations to obtain the value of absorption energies reliably and efficiently.
Keywords :
absorption; chemistry computing; learning (artificial intelligence); least squares approximations; quantum chemistry; support vector machines; LS-SVM/B3LYP/STO-3G method; absorption energies; least squares support vector machine; low level density functional theory; quantum chemical calculations; quantum mechanical calculation; Absorption; Chemicals; Chemistry; Density functional theory; Least squares methods; Machine intelligence; Machine learning; Neural networks; Quantum mechanics; Support vector machines; absorption energies; quantum mechanical calculation; support vector machines;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Computation Technology and Automation (ICICTA), 2010 International Conference on
Conference_Location :
Changsha
Print_ISBN :
978-1-4244-7279-6
Electronic_ISBN :
978-1-4244-7280-2
Type :
conf
DOI :
10.1109/ICICTA.2010.862
Filename :
5523441
Link To Document :
بازگشت