作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2010, Vol. 36 ›› Issue (21): 254-256,259. doi: 10.3969/j.issn.1000-3428.2010.21.091

• 开发研究与设计技术 • 上一篇    下一篇

基于增强学习的旅行计划推荐系统

刘小燕,陈艳丽,贾宗璞,沈记全   

  1. (河南理工大学计算机科学与技术学院,河南 焦作 454003)
  • 出版日期:2010-11-05 发布日期:2010-11-03
  • 作者简介:刘小燕(1981-),女,讲师,主研方向:智能信息系统;陈艳丽,讲师;贾宗璞、沈记全,教授
  • 基金资助:
    国家自然科学基金资助项目(60375038);河南省科技计划基金资助项目(092300410216);河南理工大学青年基金资助项目(646208)

Recommender System for Travel Plan Based on Reinforcement Learning

LIU Xiao-yan, CHEN Yan-li, JIA Zong-pu, SHEN Ji-quan   

  1. (School of Computer Science and Technology, Henan Polytechnic University, Jiaozuo 454003, China)
  • Online:2010-11-05 Published:2010-11-03

摘要: 目前旅游和观光事业通过推荐系统帮助用户进行互动式对话获得目标。已有的推荐系统尽管互动性已经增强,但仍采用互动策略,在设计阶段需要指定先验。针对该问题,提出一个普遍适用的模型,基于增强学习技术设计一种旅行会话推荐系统,描述推荐系统采用的方法,总结一些关键问题。分析结果表明,该系统可自动学习自适应交互策略,

关键词: 推荐系统, 旅行计划, 交互策略

Abstract: Tourism uses recommender systems to help user acquire their goals by using interactive dialogue. Although present recommender systems’ interactivity has been enhanced, they still use interactive method. This paper presents a universally applicable model, design a conversation recommender system by using reinforcement learning technique which can learn adaptive interactive method automatically. Describe methods recommender systems use and summer up a number of important issues. Results testify the system’s application in Austrian tourism portal.

Key words: recommender system, Travel Plan(TP), interactive strategy

中图分类号: