Adaptive Bayesian personalized ranking for heterogeneous implicit feedbacks
Adaptive Bayesian personalized ranking for heterogeneous implicit feedbacks
复制标题
异构隐式反馈的自适应贝叶斯个性化排名
DOI：
10.1016/j.knosys.2014.09.013
复制
发表时间：
2015-01-01
影响因子：
8.8
通讯作者：
Ming, Zhong
中科院分区：
文献类型：
--
作者：
Pan, Weike;Zhong, Hao;Ming, Zhong
由于隐式反馈与真实的行业问题背景密切相关，最近在推荐社区中受到了极大的关注。然而，大多数工作只利用了用户的同质隐式反馈，如用户来自购买活动的交易记录，而忽略了另一种类型的隐反馈，如来自浏览活动的检查记录。后者通常更丰富，尽管它们与高度不确定的W.r.t.用户的真实偏好。本文研究了一种新的推荐问题--异质隐式反馈(HIF)，其中最根本的挑战是考试成绩的不确定性。作为响应，我们设计了一种新的偏好学习算法，通过交易记录来学习每个不确定的考试记录的置信度。具体地说，我们推广了贝叶斯个性化排名(BPR)，这是一种具有开创性的两两学习算法，用于齐次隐式反馈，并自适应地学习置信度，因此称为自适应贝叶斯个性化排名(ABPR)。ABPR具有减少考试成绩不确定性和精确内隐反馈成对偏好学习的优点。在两个公开数据集上的实验结果表明，ABPR能够有效地利用不确定的考试记录，在各种面向排名的评价指标上能够获得比现有算法更好的推荐性能。(C)2014爱思唯尔B.V.保留所有权利。
Implicit feedbacks have recently received much attention in recommendation communities due to their close relationship with real industry problem settings. However, most works only exploit users' homogeneous implicit feedbacks such as users' transaction records from "bought" activities, and ignore the other type of implicit feedbacks like examination records from "browsed" activities. The latter are usually more abundant though they are associated with high uncertainty w.r.t. users' true preferences. In this paper, we study a new recommendation problem called heterogeneous implicit feedbacks (HIF), where the fundamental challenge is the uncertainty of the examination records. As a response, we design a novel preference learning algorithm to learn a confidence for each uncertain examination record with the help of transaction records. Specifically, we generalize Bayesian personalized ranking (BPR), a seminal pairwise learning algorithm for homogeneous implicit feedbacks, and learn the confidence adaptively, which is thus called adaptive Bayesian personalized ranking (ABPR). ABPR has the merits of uncertainty reduction on examination records and accurate pairwise preference learning on implicit feedbacks. Experimental results on two public data sets show that ABPR is able to leverage uncertain examination records effectively, and can achieve better recommendation performance than the state-of-the-art algorithm on various ranking-oriented evaluation metrics. (C) 2014 Elsevier B.V. All rights reserved.