针对个性化服务系统中用户偏好信息片面和不精确的问题,提出了综合用户网络使用行为和用户评价反馈信息的服务偏好挖掘方法,研究了用户服务偏好的表示模型、初始化和更新方法,最终挖掘出用户更加完整和精确的服务偏好。结果表明,该方法是可行和有效的。
Aimed at the problem that users'preference information in personalized service system is incomplete and inaccurate , this paper proposed a method of mining service preference that combined users 'network behavior with evaluation feedback .The rep-resentation, initialization and update of the users'service preference model were studied .And a more complete and accurate users'service preference information was eventually gained .The experimental result shows that the method is feasible and effective .