深度神经网络在化学中的应用研究  

Application of Deep Neural Network in Chemistry

作　　者：秦琦枫 曾斌[1] 刘思莹[1] Qin Qi-feng;Zeng Bin;Liu Si-ying(East China University of Technology,Jiangxi Nanchang 330013)

机构地区：[1]东华理工大学,江西南昌330013

出　　处：《江西化工》2018年第3期1-5,共5页Jiangxi Chemical Industry

摘　　要：深度神经网络在机器视觉、语音识别以及自然语言处理等领域取得了令人瞩目的成果,在改善相关问题处理结果的同时大幅提升了解决效率,是当下人工智能快速发展的基石。在化学中引入神经网络的解决方案,能够有效的提升化学信息处理的智能化水平。国内外研究学者已经将神经网络应用在一些化学问题的处理中,如:化合物结构与性质的定量关系研究、有机反应产物预测、化合物属性预测等。本文着重介绍了现有深度神经网络模型的基本框架,概述相关研究的进展,并针对深度神经网络在化学中的应用进行展望。Deep neural networks have achieved remarkable results in the fields of machine vision,speech recognition,and natural language processing. They have greatly improved the efficiency of solving problems while improving the results of relevant problems. This is the cornerstone of the rapid development of artificial intelligence at the moment. The introduction of neural network solutions in chemistry can effectively increase the level of intelligence in chemical information processing.Researchers at home and abroad have applied neural networks to the treatment of some chemical problems,such as the quantitative relationship between the structure and properties of compounds,the prediction of organic reaction products,and the prediction of compound properties. This article focuses on the basic framework of the existing deep neural network model,summarizes the progress of related research,and looks into the application of deep neural networks in chemistry.

关 键 词：深度神经网络 定量关系 化合物属性 

分 类 号：O6-04[理学—化学]