In this “big-data" era, vast amount of continuously arriving data can be found in various fields, such as sensor networks, network management, web and financial applications. To process such data, algorithms are usually challenged by its complex structure and high volume. Representation learning facilitates the data operation by providing a condensed description of patterns underlying the data. In this talk, we introduce three representations of temporal and streaming data. The first one is representation learning from human movement trajectories. The second one is online Piecewise Linear Representation (PLR), which constructs a number of consecutive line segments to approximate the data stream. The third one is online density estimator, which builds models for characterizing the dynamic density of the data stream. Knowledge discovery from new representations will be introduced and applications to different problems will be demonstrated.
Dr. Xiangliang Zhang is an Assistant Professor of Computer Science and directs the Machine Intelligence and kNowledge Engineering (http://mine.kaust.edu.sa) group at KAUST, Saudi Arabia. Prior to joining KAUST, she was a European ERCIM research fellow in Norwegian University of Science and Technology (NTNU), Norway, in 2010. She earned her Ph.D. degree in computer science from INRIA-Universite Paris-Sud, France, in July 2010. She received M.S. and B.S. degrees from Xi’an Jiaotong University, China, in 2006 and 2003, respectively. Dr. Zhang's research mainly focuses on learning from complex and large-scale streaming data. Dr. Zhang has authored or co-authored over 60 research papers in referred international journals and conference proceedings, including TKDE, SIGKDD, VLDB J, AAAI, IJCAI, ICDM, ECML/PKDD, CIKM, InfoCom etc. She is the reviewer of TKDE, TKDD, VLDB J, Information Science, DMKD and KAIS. She serves on the Program Committee for premier conferences like SIGKDD 2014-2016, ICDM 2013-2016, AAAI 2016, WSDM 2015 etc.