Contacts

Dirk Hovy

Department of Computing Sciences
Associate Professor
Download CV
Twitter
Office Hours and Appointments

Bio

I am a professor working on natural language processing and computational social science. Previously, I was faculty and postdoc in Copenhagen, got a PhD from USC, and a master's in sociolinguistics in Germany.

I am also the scientific director of BIDSA’s Data and Marketing Insights (DMI) research unit, and head of the MilaNLP lab. I have organized a conference (EMNLP 2017) and various workshops (on abusive language, ethics in NLP, and computational social science).

Outside of work, I enjoy cooking, leather-crafting, and picking up heavy object to put them back down. 
 

About

I received a 2020 ERC Starting Grant to explore the effect of sociodemographic variables on NLP models, and to integrate the two. 

 

Research Interests

I work in natural language processing and computational social science, and am interested in modeling what language can tell us about society, and what computers can tell us about language. I also work on ethics and fairness in NLP and AI.

Technique-wise, I am interested in large language models, transfer learning, multitask learning, reinforcement learning, and adversarial learning.

working papers

Publications

What's in a preposition? Dimensions of sense disambiguation for an interesting word class
Proceedings of Coling, 2010
Visualizing regional language variation across Europe on Twitter
Handbook of the changing world language map, 2019
Lörres, Möppes, and the Swiss. (Re)Discovering regional patterns in anonymous social media data
JOURNAL OF LINGUISTIC GEOGRAPHY, 2019
Personality traits on Twitter - or - how to get 1.500 personality tests in a week
6th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis WASSA 2015. Workshop Proceedings, 2015
Comparing Bayesian models of annotation
TRANSACTIONS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, 2018
Identifying linguistic areas for geolocation
Proceedings of the 2019 EMNLP Workshop W-NUT: The 5th Workshop on Noisy User-generated Text, 2019
Geolocation with attention-based multitask learning models
Proceedings of the 2019 EMNLP Workshop W-NUT: The 5th Workshop on Noisy User-generated Text, 2019
Increasing in-class similarity by retrofitting embeddings with demographic information
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, 2018
Hey Siri. Ok Google. Alexa: a topic modeling of user reviews for smart speakers
Proceedings of the 2019 EMNLP Workshop W-NUT: The 5th Workshop on Noisy User-generated Text, 2019
Dense node representation for geolocation
Proceedings of the 2019 EMNLP Workshop W-NUT: The 5th Workshop on Noisy User-generated Text, 2019
Computerlingvistik. Metoder til visualisering af regional variation i sociale medier
Sociale Medier Og Sprog, 2018
Women’s syntactic resilience and men’s grammatical luck: gender-bias in part-of-speech tagging and dependency parsing
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 2019
Hateful symbols or hateful people? Predictive features for hate speech detection on Twitter
The 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Proceedings of the Student Research Workshop, 2016
Models and training for unsupervised preposition sense disambiguation
Proceedings of ACL, 2011
The social and the neural network: how to make natural language processing about people again
Proceedings of the Second Workshop on Computational Modeling of People’s Opinions, Personality, and Emotions in Social Media, 2018
More or less supervised supersense tagging of Twitter
Proceedings of the Third Joint Conference on Lexical and Computational Semantics, 2014
Linguistically debatable or just plain wrong?
Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, 2014
Challenges of studying and processing dialects in social media
ACL 2015 Workshop on Noisy User-generated Text. Proceedings of the Workshop, 2015
User review sites as a resource for large-scale sociolinguistic studies
Proceedings of the 24th International Conference on World Wide Web, 2015
Mining for unambiguous instances to adapt POS taggers to new domains
Proceedings of NAACL, 2015

Teaching

20879
LANGUAGE TECHNOLOGY
40996
COMPUTATIONAL METHODS FOR SOCIAL AND POLITICAL SCIENCE
Past years teaching