BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//hacksw/handcal//NONSGML v1.0//EN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20260511T033936Z
DESCRIPTION:Click for Latest Location Information: http://edw2018.dataversi
 ty.net/sessionPop.cfm?confid=121&proposalid=9368\nFor the past year, news h
 as been the hottest item in the news – from fake news to inadvertentl
 y inappropriate news alongside ads to problems of bias with Facebook’
 s trending news. Organizations have been trying a number of techniques to f
 lag news with any of these issues but with limited success. One of the main
  reasons is that these techniques lack the ability to deal with the meaning
  expressed within those news stories. This talk describes the use of a numb
 er of text analytics approaches that range from comparing patterns of words
  that can distinguish fake news from real to categorizing the content of ne
 ws more accurately and consistently and thus flag inappropriate news. We wi
 ll share the results of two studies that demonstrated the efficacy of a num
 ber of approaches but showed even more clearly the effectiveness of combini
 ng machine learning-pattern matching with categorization rules. It might no
 t restore civility to politics but it might help make the news less a force
  for division.
DTSTART:20180425T120000
SUMMARY:Fake News – Using Text Analytics to Tag Fakes
DTEND:20180425T125959
LOCATION: See Description
END:VEVENT
END:VCALENDAR