改进的Web文本自适应过滤策略  

Improvement of Web Text Filtering Strategy

作　　者：沈凤仙[1] 朱巧明[2] 刘粉香[1] 

机构地区：[1]三江学院计算机基础教学部,江苏南京210012 [2]江苏省计算机信息处理技术重点实验室,江苏苏州215006

出　　处：《计算机与现代化》2010年第9期48-52,共5页Computer and Modernization

摘　　要：面对实时网络信息过滤的新挑战,自适应信息过滤基本上能够解决问题。针对现有自适应系统的不足,本文提出提高模板准确性的学习和过滤阈值优化的新方法。改进的过滤策略过滤初期采用SVM算法,中后期采用改进的自适应模板过滤法。模板的更新采用改进的模板系数调整策略,并引入特征衰减因子来提高过滤的准确率。该系统运行于一个校园网关上,取得了较好的结果。In order to meet needs of the real-time information filtering,experts put forward adaptive information filtering method.But there are still some difficulties in training and adaptive learning.This paper proposes a new method to improve the accuracy of the adaptive filtering model and optimize the filtering threshold.The filtering strategy uses SVM at initial filtering stages and uses the improved adaptive template-based algorithm in latter stages.In order to update profile,it uses the improved coefficient adjustment strategy,and uses the feature attenuation factor.Running on a gateway in the campus,this system has achieved good results.

关 键 词：网页过滤 自适应信息过滤 语义倾向 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]