首页> 外文会议>International Conference on Fuzzy System and Data Mining >An Information Recommendation Method Based on User Interest Model
【24h】

An Information Recommendation Method Based on User Interest Model

机译:一种基于用户兴趣模型的信息推荐方法

获取原文

摘要

Most of the existing works on travel recommendations are based on check-in data or photos. From a distinct point of view, the present research pays attention to network travelogues that contain a large number of users' evaluation information on tourism items. By analyzing the sentiment inclinations, demographic information and user's web access actions, a hybrid user interest model based on ontology is established. When a query is submitted, neighbors who have similar sentiment inclinations are found by this model, and then the interest degree of tourism items that are located in the same category of target query item can be predicted according to the user's historical behavior. The results show that our hybrid ontology model can make tourism item recommendations more effectively than the standard or pure ontology models, and better solve the cold start problem.
机译:旅行建议上的大多数现有工作基于登记数据或照片。从独特的观点来看,目前的研究会注意包含关于旅游项目大量用户评估信息的网络旅行。通过分析情绪倾向,人口统计信息和用户的Web访问动作,建立了基于本体的混合用户兴趣模型。当提交查询时,该模型发现具有类似情绪倾斜的邻居,然后可以根据用户的历史行为预测位于同一类别查询项中的旅游项目的兴趣度。结果表明,我们的杂交本体模型可以比标准或纯本体模型更有效地制作旅游项目的建议,更好地解决了冷启动问题。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号