William L. Hamilton

I use computational models to analyze large social systems, with an emphasis on extracting linguistic signals that reveal the dynamics of social behavior, and I develop machine learning approaches that allow researchers to learn useful, predictive representations of high-dimensional social data.

2017
Community Identity and User Engagement in a Multi-Community Landscape
Justine Zhang*, William L. Hamilton*, Cristian Danescu-Niculescu-Mizil, Jure Leskovec, Dan Jurafsky.
*Equal contributions.
Proceedings of ICWSM. 2017 (to appear).
Loyalty in Online Communities
William L. Hamilton*, Justine Zhang*, Cristian Danescu-Niculescu-Mizil, Jure Leskovec, Dan Jurafsky.
*Equal contributions.
Proceedings of ICWSM (short paper). 2017 (to appear).
2016
Inducing Domain-Specific Sentiment Lexicons from Unlabeled Corpora
William L. Hamilton, Kevin Clark, Jure Leskovec, Dan Jurafsky.
Proceedings of EMNLP. 2016.
pdf     project website (code+data)
Cultural Shift or Linguistic Drift? Comparing Two Computational Models of Semantic Change
William L. Hamilton, Jure Leskovec, Dan Jurafsky.
Proceedings of EMNLP. 2016.
pdf     project website (code+data)
Learning Linguistic Descriptors of User Roles in Online Communities
Alex Wang, William L. Hamilton, Jure Leskovec.
EMNLP Workshop on Computational Social Science (NLP+CSS). 2016.
pdf
Diachronic Word Embeddings Reveal Statistical Laws of Semantic Change
William L. Hamilton, Jure Leskovec, Dan Jurafsky.
Proceedings of ACL. 2016.
pdf     project website (code+data)
Predicting the Rise and Fall of Scientific Topics from Trends in their Rhetorical Framing
Vinodkumar Prabhakaran, William L. Hamilton, Dan McFarland, Dan Jurafsky.
Proceedings of ACL. 2016.
pdf
2014
Compressed Predictive State Representation: An Efficient Moment-Method for Sequence Prediction and Sequential Decision Making
William L. Hamilton
MSc Thesis. McGill University.
Canadian AI Association (CAIAC) 2014 MSc Thesis Award
pdf
Methods of Moments for Learning Stochastic Languages: Unified Presentation and Empirical Comparison
Borja Balle*, William L. Hamilton*, Joelle Pineau.
*Equal contributions.
Proceedings of ICML. 2014.
pdf
Efficient Learning and Planning with Compressed Predictive States
William L. Hamilton, Mahdi Milani Fard, Joelle Pineau.
Journal of Machine Learning Research (JMLR). 2014.
pdf  code
2013
Modelling Sparse Dynamical Systems with Compressed Predictive State Representations
William L. Hamilton, Mahdi Milani Fard, Joelle Pineau.
Proceedings of ICML. 2013.
pdf  code

I am a PhD student in computer science at Stanford University, working jointly in the NLP and SNAP groups, which I joined in 2015. My interests lie at the intersection of network science, natural language processing, and computational social science. I am advised by Dan Jurafsky and Jure Leskovec. My PhD is supported by a SAP Stanford Graduate Fellowship and a Canadian NSERC Grant.

Prior to coming to Stanford, I completed a BSc and MSc at McGill University, where I studied computer science (with an emphasis on theoretical machine learning) in the Reasoning and Learning Lab under the supervision of Joelle Pineau.

During the summers of 2013 and 2014 I interned at Amazon as a software development engineer and research scientist. My work involved designing and implementing machine learning (specifically, time-series prediction) algorithms.

I grew up in the prairies of Canada (Saskatoon, Saskatchewan to be exact), so if I ever pronounce something in a strange manner, you can blame it on that.


Stanford University Stanford NLP SNAP

PhD Candidate
Computer Science, Stanford University

wleif(at)stanford.edu

Gates 450
Stanford CA 94305


CV
Google Scholar


Many thanks to David Jurgens for the site template/inspiration