报告题目：Personalized Prioritization for Broadcast Emails——From Active Learning to Cross-domain Recommendation
报告人：Dr. Beidou Wang
Abstract: With email overload becoming a billion-level drag on the economy, personalized email prioritization is of urgent need to help predict the importance level of an email. Despite lots of previous effort on the topic, broadcast email, an important type of emails with its unique challenges and intriguing opportunities, has been overlooked. The most salient opportunity lies in that effective collaborative filtering can be exploited due to thousands of receivers of a typical broadcast email. However, every broadcast email is completely ``cold''.
To solve the cold start problem, we propose the first framework for broadcast email prioritization by designing a novel active learning model that considers the collaborative filtering, implicit feedback and time sensitive responsiveness features of broadcast emails.
However, even with the active learning framework, it is still very costly to obtain users' preference feedback. Fortunately, there exist up to million-level broadcast mailing lists in a real life email system. Similar mailing lists can provide useful extra information for broadcast email prioritization in a target mailing list. How to mine such useful extra information is a challenging problem that has never been touched.
We then propose the first broadcast email prioritization framework considering large numbers of mailing lists by formulating this problem as a cross domain recommendation problem. An optimization framework is proposed to select the optimal set of source domains considering multiple criteria including overlap of users, feedback pattern similarity, and coverage of users. Our method is thoroughly evaluated on a real world industrial dataset from Samsung Electronics and is proved highly effective and outperforms all the baselines.
This talk will not only cover the interesting broadcast email prioritization problem but also introduce the active learning and cross domain recommendation framework behind it in details.
Speaker bio: Beidou Wang is a PhD. from the duo-PhD degree program of Simon Fraser University, Canada and Zhejiang University, China, who now lives and studies in Vancouver, BC, Canada. He received his B.S. degree from Zhejiang University (Graduating with Honors). He worked on social network related recommendation problems, active learning, cross-domain recommendation and recommendation for Internet of Things, with multiple researched papers published on related conferences like KDD, WWW, and AAAI. He currently works as a research assistant at Simon Fraser University Canada and a data scientist at Samsung Research Canada. He is a reviewer/external reviewer for multiple conferences including WWW, AAAI, NIPS and SDM.
He is also the founder of IKAN, the NO. 1 TV show schedule management APP in China, with over half million APP users and million-level seed investment from Qihu360, Netease and Haoan Foundation.