摘要
探讨用遗传算法训练神经网络，为苯乙酰胺类化合物的ＱＳＡＲ建模，效果良好．神经网络可以反映复杂的构效关系，而引入遗传算法又有助于多层前传网在训练过程中跳出局部最小点，使收敛速度大大提高，并在预报精度上有显著改善．
In this paper, a neural network based on genetic algorithms was proposed for a QSAR model building of herbicidal N (1 methyl 1 phenylethyl)phenylacetamides. Since neural networks can express complicated structure activity relationships, and the combination of Genetic Algorithms can help the networks to jump out of the local optimal points, thus, it speeds up the covergence of the training. At the same time, it's prediction precision has been notably raised.
出处
《高等学校化学学报》
SCIE
EI
CAS
CSCD
北大核心
1998年第6期871-875,共5页
Chemical Journal of Chinese Universities
基金
浙江省自然科学基金
