BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:From Sentiment Analysis to Topic Discovery
DTSTART:20140602T100000
DTSTAMP:20260410T125159Z
UID:53e15252e7e08b71d2321cae76102054716890adbe5a3f60b129eeb7
CATEGORIES:Conferences - Seminars
DESCRIPTION:Bing LIU\, University of Illinois at Chicago\nSentiment analys
 is (SA) or opinion mining is the computational study of people's opinions\
 , sentiments\, attitudes\, and emotions expressed in written language. It 
 is one of the most active research areas in natural language processing (N
 LP) and data mining. The popularity of SA is mainly due to two factors: se
 emingly unlimited applications and many challenging research problems. SA 
 can be regarded as a semantic analysis problem\, but it is also highly tar
 geted and bounded because a SA system does not need to fully "understand" 
 each sentence or document\, but only needs to comprehend some aspects of i
 t\, e.g.\, positive/negative opinions and their targets. SA\, however\, do
 es not seem to be just a sub-problem of NLP. It is more like a mini versio
 n of full NLP because SA touches upon every core issue of NLP.\nHowever\, 
 due to the targeted and bounded nature of SA\, it allows us to perform dee
 per language analyses to gain much better insights into NLP than in the ge
 neral setting because the complexity of the general setting of NLP is simp
 ly too overwhelming. In this talk\, I will first define SA and discuss the
  current state of the art\, and then go into details to discuss one recent
  study that aims to solve a SA problem but also contributes to machine lea
 rning in the areas of topic modeling\, lifelong learning\, and big data.
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR