【海韵讲座】2020年第1期-Towards Optimal Outdoor Advertising: When Impression Counts Meet Moving Trajectories
发布时间:2019-12-31 点击:


海韵讲座第1


报告题目Towards Optimal Outdoor Advertising: When Impression Counts Meet Moving Trajectories

主讲人鲍芝峰 澳大利亚皇家墨尔本理工大学

时间202012号上午1000

地点:海韵园行政楼A306

摘要:

In this talk I will present our work as titled which received the KDD 2019 Best Paper Award Runnerup. In particular, we studied the problem of maximizing the influence for outdoor advertising when users’ mobility and the impression counts to such users are taken into consideration. In line with the advertising consumer behavior studies, we adopted the logistic function to take impression counts of a billboard to a user trajectory into consideration when defining the influence measurement. We proved that this problem is NP-hard to approximate and propose a range of solutions to seek a controllable tradeoff between effectiveness and efficiency, with non-trivial theoretical guarantees. We conducted extensive experiments over real-life trajectory data and outdoor ads data to evaluate the performance of the above.

报告人简介:

鲍芝峰,澳大利亚皇家墨尔本理工大学(RMIT)计算机学院副教授,墨尔本大学荣誉研究员。鲍博士于2011年在新加坡国立大学(NUS)计算机学院获得博士学位,是该学院唯一的最佳博士论文奖获得者,也是新加坡信息发展局(InfoComm Development Authority)金奖获得者,并曾两次获得Google Faculty Research Award奖项。鲍博士研究方向主要包括时空数据分析、数据集成和数据可视化等,担任PVLDB 2021副主编,WSDM19 CUPDASFAA17workshop track)、ER18demo track)的PC联合主席,以及VLDB17-20SIGMOD18SIGIR15-19ICDE16-20WWW 18-19等顶级会议的PC成员。鲍博士获得了KDD 2019 Best Paper Award  RunnerupDASFAA17 Best Student Paper Runnerup5项最佳论文奖,以及KDD 2018ICDE 2009CIKM 20145项最佳论文提名。自2015年以来,他作为负责人承担了来自澳大利亚研究委员会(Australian Research Council)、CSIRO、谷歌(Google)等机构提供的累计150多万澳元的科研项目。截止目前,鲍博士在国际顶级期刊和会议(CCF-A类,如TKDE, PVLDB, SIGMOD, SIGKDD, SIGIR, VLDB, ICDE, WWW等),已累计发表30余篇高水平研究论文。

邀请人:软件工程系 王晓黎