基于改进SVM的网页过滤系统研究  

Researh of Webpage fltering system based on improved SVM

作　　者：李兆翠[1] 朱振方[2] 李颖[1] 

机构地区：[1]山东协和学院计算机学院 [2]山东交通学院信息科学与电气工程学院,山东济南250100

出　　处：《软件导刊》2016年第2期159-161,共3页Software Guide

基　　金：山东协和学院校级科技项目(XHXY201417)

摘　　要：对比分析了几种常见的网络文本分类方法。其中,支持向量机具有较高的分类准确率。提出支持向量机改进算法,将多项式核函数和径向基核函数加权组合成混合核函数。该核函数克服了支持向量机中单个核函数的局限性,可兼顾算法的学习能力和泛化能力。对建立的网页过滤模型进行仿真实验,证明改进的SVM网页过滤模型能提高过滤的准确率和效率。In this paper, several common network text classification methods are analyzed, and then the support vector machine algorithm is found that it has higher classification accuracy. We propose improved support vector machine algorithm,combining polynomial kernel and radial basis function to a mixed weighted kernels. The kernel overcome the limitations of a single kernel of SVM, the algorithm can be both learning ability and generalization ability. Simulation experiments arecarried out to establish the model of Web filtering, which can improve the accuracy and efficiency of the SVM web page filtering model.

关 键 词：SVM 网页过滤 混合核函数 文本分类 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]