论文部分内容阅读
本文采用简单的化学基团描述符来表征化学物质的结构、以底物的最大比去除速率表征生物可降解性的大小,运用自编的人工神经网络对芳香族化合物的生物可降解性进行了研究.试验过程中,按芳香族化合物最大比底物去除速率的大小将其生物可降解性分为四组:不降解、难降解、可降解、易降解.随机抽取10%的有机化合物作为预测集,余下的作为训练集,共进行了四次预测试验,正确预测率分别达80%,80%,4O%,80%,这显示了人工神经网络对芳香族化合物的生物可降解性有较好的预测性能.
In this paper, simple chemical group descriptors were used to characterize the structure of chemical substances. The biodegradability was characterized by the maximum specific surface area removal ratio. The biodegradability of aromatic compounds was determined by self-made artificial neural network In the course of the experiment, the biodegradability of aromatic compounds was divided into four groups according to the maximum specific substrate removal rate: non-degradable, degradable, degradable, and degradable.A random 10% organic compound was selected as the prediction Set, and the rest as a training set, a total of four prediction tests were conducted with the correct prediction rates of 80%, 80%, 40% and 80%, respectively. This shows that artificial neural networks have more biodegradability for aromatic compounds Good predictive performance.