报告题目：Fast Prediction of YouTube Video Popularity through a Lifetime Aware Approach

报 告 人: 颜至声教授

主 持 人：郭文忠教授

报告时间：2018年7月17日 上午10:00-17:30

报告地点：福州大学数学与计算机科学学院2号楼311  

报告摘要：

Online content popularity prediction provides substantial value to a broad range of applications in the end-to-end social media systems, from network resource allocation to targeted advertising. While using historical popularity can predict the near-term popularity with a reasonable accuracy, the bursty nature of online content popularity evolution makes it difficult to capture the correlation between historical data and future data in the long term. Although various existing efforts have been made toward long-term prediction, they need to accumulate a long enough historical data before the prediction and their model assumptions cannot be applied to the complex YouTube networks with inherent unpredictability. In this talk, we aim to make the case for fast prediction of long-term video popularity as soon as after the video is uploaded in YouTube networks. We introduce LARM, a Lifetime Aware Regression Model to compensate the insufficiency of historical data without assumptions of network structure. Our results of LARM have shown up to 20% prediction error reduction in two realistic YouTube datasets.

报告人简介：

颜至声教授现任美国佐治亚州立大学计算机系的助理教授。他于2017年取得美国纽约州立大学布法罗分校计算机科学与工程系博士学位，并于2017年在美国斯坦福大学担任访问研究员。他的研究重点是物联网中的视觉计算。他的博士论文在国际上首次建立了视觉计算理论与移动系统之间的联系，获得了纽约州立大学布法罗分校CSE Best Dissertation Award。2016年，他被授予纽约州立大学布法罗分校最富盛名的David M. Benensson纪念奖学金。颜教授的跨学科研究已在MobiCom、MM、CIKM等国际顶级会议和TMC、TCSVT、TMM等顶级期刊上发表了24篇文章。他目前还担任ACM MM 2018会议程序委员会领域主席，以及多个国际会议程序委员会委员（NOSSDAV 2018, ICNC 2018, WearSys 2018, IPCCC 2018）。