蛋白质二级结构预测的多核学习方法  

Multi-kernel learning method for predicting protein secondary structures

作　　者：连云涓[1] 熊惠霖[1] 

机构地区：[1]上海交通大学自动化系,上海200240

出　　处：《计算机应用》2013年第A01期43-45,共3页journal of Computer Applications

摘　　要：核方法应用于蛋白质二级结构预测已经有数年的历史。不过,基于单核的分类器只能将所有的特征一并处理,而非区别对待以利用其特性。为了将不同的特征融合,提出一种多核学习的方法,以实验方法对比单核和多核学习在蛋白质二级结构预测效果上的差别,并提出了多核学习方法可改进现有的核分类器。Kernel methods have been applied in protein secondary structure prediction for years. But unfortunately, classifiers with one single kernel can get use of only one kind of property from protein sequence. For the sake of mixing the features form different platforms together, a multi-kernel learning framework which merges several protein properties together was performed to predict secondary structures, and experiments were carried out to show the advance.

关 键 词：多核学习 核方法 蛋白质二级结构预测 支持向量机 生物信息学 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]