摘要
对Logistic回归的输出结果通过概率分析划分为四个连续的区间,计算各个区间内训练样本的正确分类频率,由此将Logistic回归与支持向量机对样本的输出结果进行比较,构造了一种集成判别规则的二分类算法。实证分析表明提出的集成算法具有较好的分类效果。
By probability values,the output of quency of classification in each internal can be Logistic regression can be divided into four calculated.Based on Logistic regression and an integrated binary-class classification algorithm is proposed.The validity of BLR-SVM is several UCI datasets. continuous intervals,and the freSupport Vector Machine (SVM), illustrated by numerical results on
出处
《计算机工程与应用》
CSCD
北大核心
2011年第29期149-150,157,共3页
Computer Engineering and Applications
基金
中央高校基本科研业务费资助(No.CDJXS10100026)