BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//University of Liverpool Computer Science Seminar System//v2//EN
BEGIN:VEVENT
DTSTAMP:20260624T162655Z
UID:Seminar-networks-1093@lxserverM.csc.liv.ac.uk
ORGANIZER:CN=Giorgos Christodoulou:MAILTO:G.Christodoulou@liverpool.ac.uk
DTSTART:20210715T120000
DTEND:20210715T130000
SUMMARY:Networks and Distributed Computing Series
DESCRIPTION:Conrado Martinez : Some Probabilistic and Combinatorial Techniques in Data Stream Analysis\n\nI will give in this talk a brief account of some elegant and mathematically sound techniques\nand algorithms that provide practical solutions to fundamental problems in data streaming. These include cardinality estimation (counting the number of distinct elements in a stream), drawing random samples, finding heavy hitters (elements of high relative frequency) and top-$k$ most frequent elements, estimating the similarity of two streams, etc.\n \nIn the talk I will focus in one or two of the problems above, give a few well-known examples (the list from which we can select is huge, for example, Probabilistic Counting, LogLog and HyperLogLog, KMV,  SpaceSaving, Frequent, StickySampling, Adaptive Sampling, \ldots) and I will also present some of the results of my joint research with A. Helmi, J. Lumbroso, J. Montes, G. Solera, A. Viola and J. Wang, in recent years.\n\nhttps://www.csc.liv.ac.uk/research/seminars/abstract.php?id=1093
LOCATION:
END:VEVENT
END:VCALENDAR