BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:NLIP Seminar Series
SUMMARY:Mining the Social Web: A series of statistical NLP
  case studies - Vasileios Lampos
DTSTART;TZID=Europe/London:20141205T120000
DTEND;TZID=Europe/London:20141205T130000
UID:TALK54049AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/54049
DESCRIPTION:[ "Slides":http://www.lampos.net/sites/default/fil
 es/slides/Lampos2014SNLP_on_Twitter.pdf ]\n\n*Abst
 ract*\n\nOver the past few years user-generated co
 ntent has been the centre of various research effo
 rts in the domain of statistical natural language 
 processing. In particular\, the open nature of the
  microblogging platform of Twitter provided the op
 portunity for various appealing ideas to be evalua
 ted. Based on the hypothesis that this online stre
 am of content should represent at least a biased f
 raction of real-world situations or opinions\, we 
 have proposed core algorithms for nowcasting the r
 ate of an infectious disease\, such as influenza\,
  or even a natural phenomenon like rainfall rates 
 [1\,2]. A simplified emotion analysis on a longitu
 dinal set of tweets revealed interesting patterns\
 , including signs of rising anger and fear before 
 the UK riots in August\, 2011 [3]. By extending li
 near text regression approaches to embed user rele
 vance\, we proposed a family of bilinear regularis
 ed regression models\, which found application in 
 the approximation of voting intention trends [4]. 
 Finally\, we attempted to reverse the previous mod
 elling principle to look into how various user att
 ributes or behaviours may influence a generic noti
 on of user-impact [5].\n\n*References*\n# Lampos a
 nd Cristianini. Tracking the flu pandemic by monit
 oring the Social Web\, Cognitive Information Proce
 ssing '10.\n# Lampos and Cristianini. Nowcasting E
 vents from the Social Web with Statistical Learnin
 g\, ACM TIST (2012).\n# Lansdall-Welfare\, Lampos 
 and Cristianini. Effects of the Recession on Publi
 c Mood in the UK\, WWW '12.\n# Lampos\, Preotiuc-P
 ietro and Cohn. A user-centric model of voting int
 ention from Social Media\, ACL '13.\n# Lampos\, Al
 etras\, Preotiuc-Pietro and Cohn. Predicting and C
 haracterising User Impact on Twitter\, EACL '14.\n
LOCATION:FW26\, Computer Laboratory
CONTACT:Tamara Polajnar
END:VEVENT
END:VCALENDAR