首页> 外国专利> NEW METHOD FOR THE DEVELOPMENT OF REGRESSION MT-QSAR AND MTK-QSAR MODELS BY APPLYING MODULAR NEURAL NETWORK

NEW METHOD FOR THE DEVELOPMENT OF REGRESSION MT-QSAR AND MTK-QSAR MODELS BY APPLYING MODULAR NEURAL NETWORK

机译:应用模块化神经网络开发回归MT-QSAR和MTK-QSAR模型的新方法

摘要

New method for the development of regression mt-QSAR and mtk-QSAR models, by applying modular neural network, by which the analysis of the relationship of the structure and activity of the compounds simultaneously for multiple biological species and for the different standards of measurements of biological effects is carried out. Modular neural network consists of two layers of modules, whereby the input layer is made of independently trained classification neural networks, where each input module is separate mt-QSAR/mtk-QSAR model, and the output layer is made of one module (by type a regression neural network), which performs regression analysis, based on binary classification of the input modules.
机译:通过使用模块化神经网络开发回归mt-QSAR和mtk-QSAR模型的新方法,通过该方法可以同时分析多种生物物种和不同测量标准的化合物的结构与活性之间的关系产生生物学效应。模块化神经网络由两层模块组成,输入层由经过独立训练的分类神经网络组成,其中每个输入模块都是独立的mt-QSAR / mtk-QSAR模型,输出层由一个模块组成(按类型回归神经网络),基于输入模块的二进制分类执行回归分析。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号