基于SVM和信息增益的属性选择算法研究

该文提出了一种新的属性选择的算法,即基于信息增益和支持向量机递归属性消除的属性选择算法。该算法保留了支持向量机的高精确性、高维健壮等优点,并通过将信息增益与其结合,克服了支持向量机由于建模时间长导致运行缓慢的缺点。该文还提出了基于接受者操作特征曲线下面积的选择属性数目的方法,并将其应用于个人信贷信用评价中,取得了良好的效果。

The Research of SVM and Information Gain-Based Attribute Selection Algorithm

This paper has proposed a new attribute selection algorithm, namely the algorithm is based on information gain and support vector machine recursive feature elimination（SVM - RFE）. The algorithm maintains the merits of SVM - RFE, such as high accuracy and high dimensional mbustness, and by integrating information gain, solves the problem of slow operation caused by the long time spent in modeling with the method of SVM - RFE. The paper has also proposed the method to determine the number of characteristics based on the area under ROC curve （AUC） and got ideal results by applying the method to the individual credit evaluation model（ICEM）.