NLP at Georgia Tech Subash Chebolu and Jacob Hoylman Revalent - - PowerPoint PPT Presentation

nlp at georgia tech
SMART_READER_LITE
LIVE PREVIEW

NLP at Georgia Tech Subash Chebolu and Jacob Hoylman Revalent - - PowerPoint PPT Presentation

NLP at Georgia Tech Subash Chebolu and Jacob Hoylman Revalent people Faculty: Jacob Eisenstein PhD Students: Yi Yang, Umashanthi Pavalanathan, Sandeep Soni, Ian Stewart, and Yuval Pinter Jacob Eisenstein Focuses on non-standard language,


slide-1
SLIDE 1

NLP at Georgia Tech

Subash Chebolu and Jacob Hoylman

slide-2
SLIDE 2

Revalent people

Faculty: Jacob Eisenstein PhD Students: Yi Yang, Umashanthi Pavalanathan, Sandeep Soni, Ian Stewart, and Yuval Pinter

slide-3
SLIDE 3

Jacob Eisenstein

  • Focuses on non-standard language, discourse, computational social science, and statistical

machine learning

  • Has 92 publications from 1999 - present day
  • Leads the computational Linguistics laboratory at Georgia Tech
  • Assistant Professor in the School of Interactive Computing at Georgia Tech
  • Teaches Natural Language Processing, Computational Journalism, and Computational Social

Science classes

slide-4
SLIDE 4

Recent Papers

  • Y. Yang, M.-W. Chang, and J. Eisenstein. Toward socially-infused information extraction: Embedding authors, mentions, and entities. In

Proceedings of Empirical Methods for Natural Language Processing (EMNLP), November 2016.

  • V. Krishnan and J. Eisenstein. “You're Mr. Lebowski, I'm The Dude”: Inducing address term formality in signed social networks. In

NAACL, 2015.

  • R. Goel, S. Soni, N. Goyal, J. Paparrizos, H. Wallach, F. Diaz, and J. Eisenstein. The social dynamics of language change in online
  • networks. In The International Conference on Social Informatics (SocInfo), November 2016
slide-5
SLIDE 5

Socially Infused Information Extraction (2016)

  • Discerning what entity is being referred to using context
  • Retweets, Mentions, Followers
  • Eliminate ambiguity in identification
slide-6
SLIDE 6

Language change in Online Networks

  • Much of language change is brought through social media
  • Exposure and influence are critical for new language to spread
  • Phonetic spellings and abbreviations spread mostly through exposure
  • Social structures can be discerned through the spread of language
  • Geographically local ties aren’t very effective conduits for language change
slide-7
SLIDE 7
slide-8
SLIDE 8

Address Term Formality (2015)

  • Names, titles, and placeholders
  • How do these terms indicate formality of a relationship?
  • For example, Sir => First name and First + Last name => dude
  • Creating signed networks with information about formality, relative power,

etc.

slide-9
SLIDE 9

Questions?