PairFac: Event Analytics through Discriminant Tensor Factorization
ACM International Conference on Information and Knowledge Management(2016)
摘要
The study of disaster events and their impact in the urban space has been conducted primarily through manually collections and analysis of surveys, questionnaires and authority documents. While there have been increasingly rich troves of human behavioral data related to the events of interest, the ability to obtain hindsight following a disaster event has not been scaled up. In this paper, we propose a novel event analytic approach called \textit{PairFac} that can automatically discover the impact of a major event from the rich human behavioral data through discriminant tensor analysis. PairFac aims to uncover the persistent patterns across multiple interrelated aspects of urban behavior including when, where and what citizens would do in a city and at the same time identify the salient changes following a potentially impactful event. We show the effectiveness of PairFac in comparison with previous methods through extensive experiments. Further, we demonstrate the advantages of our approach through case studies with real-world traffic sensor data and social media streams surrounding the 2015 Paris terrorist attacks. Our work has both methodological and analytical contributions to the study of disaster events and their impact in urban space.
更多查看译文
关键词
Tensor Factorization,Terrorist Attacks,Event Analytics,Urban Computing
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络