dirk hovy

22. 10. 2016, 04:59 Uhr
BLOG: I had some time and analyzed the US presidential debates from a quantitative point of view. Turns out the candidates differ even beyond their messages.
mehr…

15. 09. 2016, 06:25 Uhr
NEWS: Ich bin froh und stolz seit 1. Aug ich als Associate Professor am Informatik Institut (DIKU) der Universität Kopenhagen beschäftigt zu sein! Das Ziel einer 15 Jahre langen Reise :)


“Do NOT read beauty magazines, they will only make you feel ugly!”
Baz Luhrman―Everybody’s Free (to Wear Sunscreen)

portfolio

Meine Forschung konzentriert sich auf computational sociolinguistics, den Schnittpunkt von Soziolinguistik und statistischer natürlicher Sprachverarbeitung (NLP). Ich bin am Einfluss sozialer Dimensionen auf Sprache und Software-Entwicklung interessiert, zum Beispiel: wie beeinflussen Altersunterschiede Tagging-Genauigkeit? Sprache kodiert soziodemografische Unterschiede (Alter, Geschlecht, etc.), aber aktuelle NLP Modelle gehen davon aus, dass alle Sprache homogen ist. Diese Annahme fügrt zu demographisch voreingenommene Modelle, die schlechte Ergebnisse auf Daten aus Bevölkerungsgruppen, die nicht mit den Trainingsdaten übereinstimmen erzielen. Das Ziel meiner Forschung ist es, soziolinguistische Erkenntnisse in NLP-Modelle zu integrieren. Konkret benutze ich Statistiken um zu erfassen und zu modellieren, wie die Interaktion zwischen demografischen Profilen und Sprachverwendung funktioniert (siehe hier oder hier). Die möglichen Ergebnisse der Forschung zu demografischem Bias helfen der Gesellschaft, Industrie, und Forschung. Ich interessiere mich außerdem für Semantik (die Bedeutung von Wörtern im Kontext, siehe hier, hier oder hier) und Non-Standard-Sprache (siehe hier, hier, hier oder hier).
Die folgenden Arbeiten können frei als PDF-Files heruntergeladen werden.

NB: Arbeiten in Natural Language Processing werden auf Grund der schnellen Entwicklung hautsächlich in Konferenzen publiziert. Jede Publikation wird im Doppelblind-Verfahren von mindestens drei Gutachtern reviewt. Annahmeraten sind (soweit verfügbar) angegeben. Autoren-Reihenfolge spiegelt normalerweise Arbeitsleistung und Verantwortung wieder.

2017

  1. Dirk Hovy, Shannon Spruit, Margaret Mitchell, Emily M. Bender, Michael Strube, and Hanna Wallach: Proceedings of the First ACL Workshop on Ethics in Natural Language Processing. EACL, 2017.
    [Inhalt]  [PDF]
  2. Adrian Benton, Margaret Mitchell, and Dirk Hovy: Multitask Learning for Mental Health Conditions with Limited Social Media Data. EACL, 2017.
    [Inhalt]  [PDF]

2016

  1. Astrid Ravn Skovse, Dirk Hovy, and Anders Johannsen: Mapping sense(s) of place. Sociolinguistic Symposium, 2016.
    [Inhalt]  [PDF]
  2. David Bamman, A. Seza Doğruöz, Jacob Eisenstein, Dirk Hovy, David Jurgens, Brendan O'Connor, Alice Oh, Oren Tsur, Svitlana Volkova: Proceedings of the First Workshop on NLP and Computational Social Science (NLPandCSS). EMNLP, 2016.
    [Inhalt]  [PDF]
  3. Nathan Schneider, Dirk Hovy, Anders Johannsen, and Marine Carpuat: SemEval-2016 Task 10: Detecting Minimal Semantic Units and their Meanings (DiMSUM). SemEval, 2016.
    [Inhalt]  [PDF]
  4. Dirk Hovy and Shannon L. Spruit: The Social Impact of Natural Language Processing. ACL short, 2016 (Akzeptanzrate 22%).
    [Inhalt]  [PDF]
  5. Dirk Hovy: The Enemy in Your Own Camp: How Well Can We Detect Statistically-Generated Fake Reviews - An Adversarial Study. ACL short, 2016 (Akzeptanzrate 22%).
    [Inhalt]  [PDF]
  6. Gavin Abercrombie and Dirk Hovy: Putting Sarcasm Detection into Context: The Effects of Class Imbalance and Manual Labelling on Supervised Machine Classification of Twitter Conversations. ACL SRW, 2016.
    [Inhalt]  [PDF]
  7. Zeerak Waseem and Dirk Hovy: Hateful Symbols or Hateful People? Predictive Features for Hate Speech Detection on Twitter. NAACL SRW, 2016.
    [Inhalt]  [PDF]
  8. Anna Jørgensen, Dirk Hovy, and Anders Søgaard: Learning a POS tagger for AAVE-like language. NAACL, 2016 (Akzeptanzrate 29%).
    [Inhalt]  [PDF]
  9. Rebekah Tromble and Dirk Hovy: These 6 charts show how much sexism Hillary Clinton faces on Twitter. Washington Post. Monkey Cage, 2016.
    [Inhalt]  [PDF]
  10. Dirk Hovy and Anders Johannsen: Exploring Language Variation Across Europe. A Web-based Tool for Computational Sociolinguistics. LREC, 2016.
    [Inhalt]  [PDF]

2015

  1. Lasse Borgholm, Peter Simonsen, and Dirk Hovy: The Rating Game: Sentiment Rating Reproducibility from Text. EMNLP, 2015 (Akzeptanzrate 24%).
    [Inhalt]  [PDF]
  2. Barbara Plank and Dirk Hovy: Personality Traits on Twitter -or- How to Get 1500 Personality Tests in a Week. WASSA, 2015.
    [Inhalt]  [PDF]
  3. Anders Johannsen, Dirk Hovy, and Anders Søgaard: Cross-lingual Syntactic Variation over Age and Gender. CoNLL, 2015 (Akzeptanzrate 29.9%).
    [Inhalt]  [PDF]
  4. Dirk Hovy and Anders Søgaard: Tagging Performance Correlates Negatively with Author Age. ACL short, 2015 (Akzeptanzrate 22.4%).
    [Inhalt]  [PDF]
  5. Željko Agić, Dirk Hovy, and Anders Søgaard: If All You Have Is a Bit of the Bible: Learning Part-of-Speech Taggers for Truly Low-Resource Languages. ACL short, 2015 (Akzeptanzrate 22.4%).
    [Inhalt]  [PDF]
  6. Anna Jørgensen, Dirk Hovy, and Anders Søgaard: Challenges of Studying and Processing Dialects in Social Media. ACL Workshop on Noisy User-generated Text, 2015.
    [Inhalt]  [PDF]
  7. Dirk Hovy: Demographic Factors Improve Classification Performance. ACL long, 2015 (Akzeptanzrate 25%).
    [Inhalt]  [PDF]
  8. Dirk Hovy, Barbara Plank, Héctor Martínez Alonso, and Anders Søgaard: Mining for unambiguous instances to adapt POS taggers to new domains. NAACL short, 2015 (Akzeptanzrate 22.1%).
    [Inhalt]  [PDF]
  9. Dirk Hovy, Anders Johannsen, and Anders Søgaard: User review sites as a resource for large-scale sociolinguistic studies. WWW, 2015 (Akzeptanzrate 14%).
    [Inhalt]  [PDF]

2014

  1. Barbara Plank, Dirk Hovy, Ryan McDonald, and Anders Søgaard: Adapting taggers to Twitter with (less) distant supervision. Coling, 2014 (Akzeptanzrate 31%).
    [Inhalt]  [PDF]
  2. Natalie Schluter, Anders Søgaard, Jakob Elming, Dirk Hovy, Barbara Plank, Héctor Martínez Alonso, Anders Johannsen, and Sigrid Klerke: Copenhagen-Malmö: Tree approximations of semantic parsing problems. SemEval, 2014.
    [Inhalt]  [PDF]
  3. Anders Johannsen, Dirk Hovy, Héctor Martínez Alonso, Barbara Plank, and Anders Søgaard: More or less supervised super-sense tagging of Twitter. *SEM (Best Paper), 2014.
    [Inhalt]  [PDF]
  4. Dirk Hovy: How Well can We Learn Interpretable Entity Types from Text?. ACL, 2014 (Akzeptanzrate 26%).
    [Inhalt]  [PDF]
  5. Dirk Hovy, Barbara Plank, and Anders Søgaard: Experiments with crowdsourced re-annotation of a POS tagging data set. ACL, 2014 (Akzeptanzrate 26%).
    [Inhalt]  [PDF]
  6. Barbara Plank, Dirk Hovy, and Anders Søgaard: Linguistically debatable or just plain wrong?. ACL, 2014 (Akzeptanzrate 26%).
    [Inhalt]  [PDF]
  7. Anders Søgaard, Anders Johannsen, Barbara Plank, Dirk Hovy, and Héctor Martínez Alonso: What’s in a p-value in NLP?. CoNLL, 2014.
    [Inhalt]  [PDF]
  8. Jakob Elming, Barbara Plank, and Dirk Hovy: Robust Cross-Domain Sentiment Analysis for Low-Resource Languages. ACL, WASSA, 2014.
    [Inhalt]  [PDF]
  9. Barbara Plank, Dirk Hovy, and Anders Søgaard: Learning POS taggers with inter-annotator agreement loss. EACL (Best Paper), 2014 (Akzeptanzrate 25%).
    [Inhalt]  [PDF]
  10. Dirk Hovy, Barbara Plank, and Anders Søgaard: When POS datasets don’t add up: Combatting sample bias. LREC, 2014 (Akzeptanzrate ~60%).
    [Inhalt]  [PDF]
  11. Yulia Tsvetkov, Nathan Schneider, Dirk Hovy, Archna Bhatia, Manaal Faruqui, and Chris Dyer: Augmenting English adjective senses with supersenses. LREC, 2014 (Akzeptanzrate ~60%).
    [Inhalt]  [PDF]
  12. Hege Fromreide, Dirk Hovy, and Anders Søgaard: Crowdsourcing and annotating NER for Twitter #drift. LREC, 2014 (Akzeptanzrate ~60%).
    [Inhalt]  [PDF]

2013

  1. Shashank Srivastava, Dirk Hovy: A Walk-based Semantically Enriched Tree Kernel Over Distributed Word Representations. EMNLP, 2013 (Akzeptanzrate 27%).
    [Inhalt]  [PDF]
  2. Dirk Hovy: Learning Semantic Types and Relations from Text. PhD Arbeit, 2013.
    [Inhalt]  [PDF]
  3. Dirk Hovy, Gopala Anumanchipalli, Alok Parlikar, Caroline Vaughn, Adam Lammert, Eduard Hovy and, Alan W Black: Analysis and Modeling of "Focus" in Context. Interspeech, 2013 (Akzeptanzrate 52%).
    [Inhalt]  [PDF]
  4. Dirk Hovy, Taylor Berg-Kirkpatrick, Ashish Vaswani, and Eduard Hovy: Learning Whom to Trust with MACE. NAACL HLT, 2013 (Akzeptanzrate 30%).
    [Inhalt]  [PDF]
  5. Mrinmaya Sachan, Dirk Hovy, and Eduard Hovy: Solving Electrical Networks to Incorporate Supervision in Random Walks. WWW Conference 2013, 2013 (Akzeptanzrate 15%).
    [Inhalt]  [PDF]
  6. Dirk Hovy, Shashank Srivastava, Sujay Kumar Jauhar, Mrinmaya Sachan, Kartik Goyal, Huying Li, Whitney Sanders, and Eduard Hovy: Identifying Metaphorical Word Use with Tree Kernels. NAACL HLT Meta4NLP Workshop, 2013.
    [Inhalt]  [PDF]

2012

  1. Dirk Hovy and Eduard Hovy: Exploting Partial Annotations with EM Training. NAACL HLT Workshop on the Induction of Linguistic Structure (WILS), 2012.
    [Inhalt]  [PDF]
  2. Dirk Hovy: Programming in Python for Linguists — A Gentle Introduction. Tutorial, 2012.
    [Inhalt]  [PDF]
  3. Dirk Hovy, James Fan, Alfio Gliozzo, Siddharth Patwardhan, and Chris Welty: When Did that Happen? — Linking Events and Relations to Timestamps. EACL, 2012 (Akzeptanzrate 26%).
    [Inhalt]  [PDF]

2011

  1. Stephan Gouws, Dirk Hovy, and Donald Metzler: Unsupervised Mining of Lexical Variants from Noisy Text. EMNLP Workshop on Unsupervised Learning in NLP (UNSUP 2011), 2011.
    [Inhalt]  [PDF]
  2. Dirk Hovy, Ashish Vaswani, Stephen Tratz, David Chiang, and Eduard Hovy: Models and Training for Unsupervised Preposition Sense Disambiguation. ACL HLT Short Papers, 2011 (Akzeptanzrate 25%).
    [Inhalt]  [PDF]
  3. Dirk Hovy, Chunliang Zheng, Eduard Hovy, and Anselmo Peñas: Unsupervised Discovery of Domain-Specific Knowledge from Text. ACL HLT, 2011 (Akzeptanzrate 26%).
    [Inhalt]  [PDF]

2010

  1. Dirk Hovy, Stephen Tratz, and Eduard Hovy: What’s in a Preposition? Dimensions of Sense Disambiguation for an Interesting Word Class. COLING, 2010 (Akzeptanzrate 41%).
    [Inhalt]  [PDF]
  2. Dirk Hovy: An Evening with… EM. Tutorial, 2010.
    [Inhalt]  [PDF]

2009

  1. Stephen Tratz, Dirk Hovy: Disambiguation of Preposition Sense Using Linguistically Motivated Features. NAACL HLT, 2009 (Akzeptanzrate 59%).
    [Inhalt]  [PDF]

2007

  1. Dirk Hovy: Regionalität in historischen Texten. Handout eines Vortrages an der Universität Kassel., 2007.
    [Inhalt]  [PDF]

2006

  1. Dirk Hovy: Sprachwandel in der Großstadtregion. Magisterarbeit, 2006.
    [Inhalt]  [PDF]

2005

  1. Dirk Hovy: Phonologische Untersuchung zu Vornamen im Deutschen. Hausarbeit, 2005.
    [Inhalt]  [PDF]

2004

  1. Dirk Hovy: Hedges. Hausarbeit, 2004.
    [Inhalt]  [PDF]
  2. Dirk Hovy: Collocations or Free Variation? Empirical Study. Hausarbeit, 2004.
    [Inhalt]  [PDF]

This Website is also available in English

Valid XHTML 1.0 Transitional Valid CSS!