图 书 馆
推荐游戏和媒体偏好的协作学习Collaborative Learning of Preferences for Recommending Games and Media |
|
| 课程网址: | http://videolectures.net/nipsworkshops2011_graepel_collaborative/ |
| 主讲教师: | Thore Graepel |
| 开课单位: | 微软研究 |
| 开课时间: | 2012-01-24 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | The talk is motivated by our recent work on a recommender system for games, videos, and music on Microsoft’s Xbox Live Marketplace with over 35M users. I will discuss the challenges associated with such a task including the type of data available, the nature of the user feedback data, implicit versus explicit, and the scale of the problem. I will then describe a probabilistic graphical model that combines the prediction of pairwise and listwise preferences with ideas from matrix factorisation and contentbased recommender systems to meet some of these challenges. The new model combines ideas from two other models, TrueSkill and Matchbox, which will be reviewed. TrueSkill is a model for estimating players’ skills based on outcome rankings in online games on Xbox Live, and Matchbox is a Bayesian recommender system based on mapping user/item features into a common trait space. This is joint work with Tim Salimans and Ulrich Paquet. Contributors to TrueSkill include Ralf Herbrich and Tom Minka, contributors to Matchbox include Ralf Herbrich and David Stern. |
| 关 键 词: | 推荐系统; 概率图形模型; 特征空间 |
| 课程来源: | 视频讲座网 |
| 数据采集: | 2022-12-14：chenjy |
| 最后编审: | 2023-05-11：chenjy |
| 阅读次数: | 18 |