BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Plone.org//NONSGML plone.app.event//EN
X-WR-TIMEZONE:Europe/Berlin
BEGIN:VEVENT
SUMMARY:Extreme analyses of graphs and text
DTSTART;TZID=Europe/Berlin:20220705T110000
DTEND;TZID=Europe/Berlin:20220705T120000
DTSTAMP:20260414T095405Z
UID:2e362dc2076042a8963b8b611d785023@www.informatik.uni-bonn.de
CREATED:20220701T085727Z
DESCRIPTION:Scientific talk: Extreme analyses of graphs and text\n\nThis t
 alk presents my research on extreme analyses of graphs and text. For examp
 le\, in text analysis we consider the task of extreme multi-label classifi
 cation and show that our simple WideMLP outperforms state-of-the-art graph
 -based models like TextGCN. In graph analysis\, we are analysing graphs wi
 th billions of edges. Our CIKM 2020 paper shows that graph summaries can b
 e efficiently computed in a parallel and incremental algorithm. This is im
 portant for reflecting temporal changes in web graphs. Finally\, we are co
 nsidering my most recent research on stratified k-bisimulation on very lar
 ge graphs with up to almost two billion edges.\n\n \nTeaching talk: Logist
 ic Regression: Classification\n\nLinear and non-linear regression are well
 -known approaches for modeling data distributions. In the MSc module "Data
  Mining and Machine Learning"\, we are considering logistic regression to 
 effectively solve binary classification tasks.
LAST-MODIFIED:20231214T225637Z
URL:https://www.informatik.uni-bonn.de/de/lost-found/veranstaltungen/extre
 me-analyses-of-graphs-and-text
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Berlin
X-LIC-LOCATION:Europe/Berlin
BEGIN:DAYLIGHT
DTSTART:20220327T030000
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR