期刊
WEB CONFERENCE 2019: PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE (WWW 2019)
卷 -, 期 -, 页码 1977-1987出版社
ASSOC COMPUTING MACHINERY
DOI: 10.1145/3308558.3313594
关键词
Recommender system; Repeat consumption; Temporal dynamics; Collaborative filtering; Hawkes process
资金
- Natural Science Foundation of China [61672311, 61532011]
- National Key Research and Development Program of China [2018YFC0831900]
Repeat consumption is a common scenario in daily life, such as repurchasing items and revisiting websites, and is a critical factor to be taken into consideration for recommender systems. Temporal dynamics play important roles in modeling repeat consumption. It is noteworthy that for items with distinct lifetimes, consuming tendency for the next one fluctuates differently with time. For example, users may repurchase milk weekly, but it is possible to repurchase mobile phone after a long period of time. Therefore, how to adaptively incorporate various temporal patterns of repeat consumption into a holistic recommendation model has been a new and important problem. In this paper, we propose a novel unified model with introducing Hawkes Process into Collaborative Filtering (CF). Different from most previous work which ignores various time-varying patterns of repeat consumption, the model explicitly addresses two item-specific temporal dynamics: (1) short-term effect and (2) lifetime effect, which is named as Short-Term and Life-Time Repeat Consumption (SLRC) model. SLRC learns importance of the two factors for each item dynamically by interpretable parameters. According to extensive experiments on four datasets in diverse scenarios, including two public collections, SLRC is superior to previous approaches for repeat consumption modeling. Moreover, due to the high flexibility of SLRC, various existing recommendation algorithms are shown to be easily leveraged in this model to achieve significant improvements. In addition, SLRC is good at balancing recommendation for novel items and consumed items (exploration and exploitation). We also find that the learned parameters is highly interpretable, and hence the model is able to be leveraged to discover items' lifetimes, and to distinguish different types of items such as durable and fast-moving consumer goods.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据