摘要
分析了网络信息过滤一般模型以及现有技术,研究了如何更准确地构建用户模板,提出了一种基于遗传算法的网络信息过滤系统模型,并且引入了遗传扩展操作和Boltzmann群体更新准则来改进遗传算法存在的缺点,同时给出了一种Roocchio反馈模型对用户兴趣模板进行更新和维护。实验结果表明,基于该模型设计的网络信息过滤系统能够有效实现对网络信息过滤。
The network model and general information filtering technology are analyzed. On how to build more accurate user templates and template learning algorithm is studied. A model of network information filtering system based on genetic algorithm is given, and the simulated annealing to improve the genetic algorithm is introduced. At the same time a model of user feedback Roocchio is presented to update and maintain interest template. Experiment shows that, the network information filtering system based on the model is achieved effectively filter information.
出处
《计算机工程与设计》
CSCD
北大核心
2009年第2期419-422,共4页
Computer Engineering and Design
基金
国家自然科学基金项目(60873247)
山东省自然基金项目(Y2006G20)
山东省高新自主创新专项工程基金项目(2008ZZ28)
关键词
信息过滤
遗传算法
模拟退火
反馈
扩展操作
information filtering
genetic algorithm
simulated annealing
feedback
expand operations
