arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1012.0498cs.LG
收藏

推荐系统中的概率估计

Estimating Probabilities in Recommendation Systems

Georgia Institute of Technology(佐治亚理工学院);Lawrence Livermore National Lab(劳伦斯利弗莫尔国家实验室)

Mingxuan Sun, Guy Lebanon, Paul Kidwell

摘要:

推荐系统正在成为一个重要的商业应用,具有重大的经济影响。目前流行的系统包括亚马逊的图书推荐,Netflix的电影推荐和Pandora的音乐推荐。在本文中,我们解决的问题,估计概率与推荐系统数据使用非参数核平滑。在我们的估计中,我们将缺失项解释为随机截尾观测,并利用生成函数的组合性质获得有效的计算方案。我们展示了我们的方法与几个案例研究,涉及现实世界的电影推荐数据。结果是与国家的最先进的技术,同时也提供了传统的推荐系统的范围之外的概率偏好估计。

英文摘要:

Recommendation systems are emerging as an important business application with significant economic impact. Currently popular systems include Amazon's book recommendations, Netflix's movie recommendations, and Pandora's music recommendations. In this paper we address the problem of estimating probabilities associated with recommendation system data using non-parametric kernel smoothing. In our estimation we interpret missing items as randomly censored observations and obtain efficient computation schemes using combinatorial properties of generating functions. We demonstrate our approach with several case studies involving real world movie recommendation data. The results are comparable with state-of-the-art techniques while also providing probabilistic preference estimates outside the scope of traditional recommender systems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑