一种改进的KMP算法在不良网站信息过滤中的应用  

Application of an improved KMP algorithm in bad website information filtering

作　　者：党红云[1] 蒋品群[1] 何婷婷[1] 

机构地区：[1]广西师范大学电子工程学院,广西桂林541004

出　　处：《现代电子技术》2012年第1期110-112,116,共4页Modern Electronics Technique

摘　　要：针对网络信息过滤的特点和现实中人们对网络信息纯净度的要求,提出了一种基于KMP字符串匹配算法,对不良网站信息进行过滤和相应的性能测试。在测试环境下,对100组非法网站进行过滤,得出对不良信息过滤查准率达到95%,查全率达到98%,通过对测试数据的分析和网络吞吐量的测试结果表明,该方案所设计的系统性能基本能够满足实际需要。According to the characteristics of network information filtering and people′s requirement on the degree of purity of network information in reality,a KMP（Kunth-Morris-Pratt）-based string matching algorithm is introduced to filter the negative website information and test the corresponding performance.In the test environment,100 groups of illegal websites were filtered.It is concluded that the filtering precision ratio on bad information has been reached 95% and recall ratio has been reached 98%.The analysis to the test data and the test results of network throughput show that the system performance designed by this scheme can basically meet the practical need.

关 键 词：信息过滤 KMP算法 模式匹配 网络吞吐量 

分 类 号：TP393[自动化与计算机技术—计算机应用技术] TP311[自动化与计算机技术—计算机科学与技术]