首页> 外国专利> MULTIPLE LINEAR REGRESSION-ARTIFICIAL NEURAL NETWORK HYBRID MODEL FOR PREDICTING THE MELTING HEAT OF PURE ORGANIC COMPOUNDS CAPABLE OF FORMING AN ARTIFICIAL NEURAL NETWORK OUTPUTTING THE MELTING HEAT BASED ON THE VALUES OF MOLECULAR DESCRIPTORS CONTAINED IN A MULTIPLE LINEAR REGRESSION MODEL

MULTIPLE LINEAR REGRESSION-ARTIFICIAL NEURAL NETWORK HYBRID MODEL FOR PREDICTING THE MELTING HEAT OF PURE ORGANIC COMPOUNDS CAPABLE OF FORMING AN ARTIFICIAL NEURAL NETWORK OUTPUTTING THE MELTING HEAT BASED ON THE VALUES OF MOLECULAR DESCRIPTORS CONTAINED IN A MULTIPLE LINEAR REGRESSION MODEL

机译：多线性回归-人工神经网络混合模型，用于预测纯有机化合物的熔化热，该有机化合物能够形成基于分子描述值的多元热值，从而形成人工神经网络。

页面导航

摘要
著录项
相似文献

摘要

PURPOSE: A multiple linear regression-artificial neural network(MLR-ANN) hybrid model for predicting the melting heat of pure organic compounds is provided to improve the performance of prediction.;CONSTITUTION: Molecular descriptors for the melting heat of hydrocarbon-based compounds are prepared. Experimental data is classified based on a training set and a testing set. The optimal MLR model(MLRM) for the training set is searched. Entire samples are divided into three sets, and the optimal ANN model(ANNM) is searched. If the absolute value of the predicted melting heat difference based on the optimal MLRM and the optimal ANNM is more than an over-fitting preventive reference value, the predicted melting heat based on the MLRM is adopted as the melting heat.;COPYRIGHT KIPO 2012

机译：目的：提供一种用于预测纯有机化合物熔融热的多元线性回归人工神经网络（MLR-ANN）混合模型，以提高预测性能。准备好了。根据训练集和测试集对实验数据进行分类。搜索训练集的最佳MLR模型（MLRM）。将整个样本分为三组，并搜索最优的神经网络模型（ANNM）。如果基于最佳MLRM和最佳ANNM的预测熔融热差的绝对值大于过拟合预防参考值，则将基于MLRM的预测熔融热用作熔融热。; COPYRIGHT KIPO 2012

著录项

公开/公告号KR20120085169A

专利类型
公开/公告日2012-07-31

原文格式PDF
申请/专利权人 CHEMESSEN INC.;
展开▼

申请/专利号KR20110102077
发明设计人 JUNG WON CHON;PARK TAE YUN;SUNG AE RI;KIM YANG SOO;KWON OH YUNG;JEON JEONG JAE;CHO JUN HYUK;KWON YUN KYUNG;
展开▼

申请日2011-10-06
分类号G06F19/00;G06N3/12;
国家 KR
入库时间 2022-08-21 17:09:27

相似文献

专利
外文文献
中文文献