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.