| 基于代理的用户偏好建模研究 |
| |
| 引用本文： | 黄海清,王永峰.基于代理的用户偏好建模研究[J].哈尔滨工业大学学报,2007,39(7):1163-1165. |
| |
| 作者姓名： | 黄海清 王永峰 |
| |
| 作者单位： | 1. 北京邮电大学,电信工程学院,北京,100876
2. 航天科工集团二院,北京,100854 |
| |
| 摘 要： | 基于移动代理系统,将马尔可夫判决过程和智能强化学习算法相结合,提出了在异构无线网络环境下对移动用户业务偏好进行智能建模的技术框架.为动态环境下用户需求的感知、量化和适配特征的研究提供了基本的数学描述,为解决用户体验的评价问题和业务与业务环境的适配问题提供了新的研究思路.
|
| 关 键 词： | 移动代理 用户偏好 马尔可夫判决过程 强化学习 |
| 文章编号： | 0367-6234（2007）07-1163-03 |
| 修稿时间： | 2006-03-06 |
Modeling of user preference based on agent |
| |
| HUANG Hai-qing,WANG Yong-feng.Modeling of user preference based on agent[J].Journal of Harbin Institute of Technology,2007,39(7):1163-1165. |
| |
| Authors: | HUANG Hai-qing WANG Yong-feng |
| |
| Abstract: | We present a technical architecture based on agent for user preference model involved in dynamic access to wireless network,and show the nature of the problem represented within a Markov decision process(MDP) combined with adaptive reinforcement learning algorithm.A possible candidate solution is provided for user modeling dynamically to satisfy the users' expected preference based on minimal or missing information,which is also a exploration for the evaluation of the user experience when selecting service providers. |
| |
| Keywords: | mobile agent user preference Markov decision process reinforcement learning |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
