• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Business information systems workshops >Building a Social Recommender System by Harvesting Social Relationships and Trust Scores between Users
【24h】

Building a Social Recommender System by Harvesting Social Relationships and Trust Scores between Users

机译：通过收集用户之间的社交关系和信任度来建立社交推荐系统

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Recommender systems were created to guide the user in a personalized way to interesting resources and to help users cope with the problem of information overload. A system's ability to adapt to the users' needs is based on gathering user-generated collective intelligence. In this paper, we present WSNRS, the system proposed for recommending content within social networks. The main goal of the system is to identify and filter the recently published valuable resources while taking into account the interactions and the relationships the user has within social structures. The interactions are logged and aggregated in order to determine the trust scores between users. Using the scores obtained, one can identify the types of relationship established between users; the scores will then be integrated into an adaptive global model used for recommending resources. Our approach presents several advantages over classic CF-based approaches and content-based recommendations regarding cold start, scalability and serendipitous recommendations. We will illustrate this with a case study that we made using data provided by the implementation of the system in a real online social network.
机译：创建了推荐器系统，以个性化的方式引导用户使用有趣的资源，并帮助用户应对信息过载的问题。系统适应用户需求的能力基于收集用户生成的集体情报。在本文中，我们介绍了WSNRS，该系统旨在为社交网络中的内容推荐。该系统的主要目标是识别并过滤最近发布的有价值的资源，同时考虑用户在社会结构中的互动和关系。记录并汇总交互，以确定用户之间的信任度。使用获得的分数，可以识别用户之间建立的关系的类型。然后将分数整合到用于推荐资源的自适应全局模型中。与传统的基于CF的方法和基于内容的建议有关的冷启动，可伸缩性和偶然性建议相比，我们的方法具有多个优势。我们将通过一个案例研究来说明这一点，该案例研究是使用真实的在线社交网络中系统实施提供的数据进行的。

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