| 统计方法在提高密度泛函理论准确性的研究进展 |
| |
| 引用本文： | 张家虎,王秀军.统计方法在提高密度泛函理论准确性的研究进展[J].分子科学学报,2009,25(4). |
| |
| 作者姓名： | 张家虎 王秀军 |
| |
| 作者单位： | 华南理工大学化学与化工学院,广东,广州,510640 |
| |
| 基金项目： | 广东省自然科学基金资助项目，华南理工大学基金 |
| |
| 摘 要： | 介绍了神经网络方法、线性回归分析方法和支持向量机模型的原理及其对密度泛函理论计算结果修正的研究进展.这3种统计方法在改进密度泛函理论计算结果准确性方面均有着很大的作用.最后讨论了3种方法亟待解决的问题并对其发展进行了展望.
|
| 关 键 词： | 密度泛函理论 神经网络 线性回归分析 支持向量机 |
Developments on correction model of density functional theory based on statistics methods |
| |
| ZHANG Jia-hu,WANG Xiu-jun.Developments on correction model of density functional theory based on statistics methods[J].Journal of Molecular Science,2009,25(4). |
| |
| Authors: | ZHANG Jia-hu WANG Xiu-jun |
| |
| Institution: | College of Chemistry and Chemical Engineering;South China University of Technology;Guangzhou 510640;China |
| |
| Abstract: | The mechenisms of the neural network approach,linear regression correction and support vector machine,and the corrected results of density functional theory(DFT) by these three approaches are introduced in this paper.The above three methods are very important in improving the accuracy of DFT,when compared with the experimental results.The current correction methods need improvement and the development orientation is also pointed out. |
| |
| Keywords: | density functional theory neural network linear regression support vector machine |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
