Core team:


Jake Williams

Berkeley

Department of Mathematics and Statistics, Assistant Professor


Most recent papers:

Andy Reagan, Chris Danforth, Brian Tivnan, Jake Williams, Peter Sheridan Dodds. Sentiment analysis methods for understanding large-scale texts: A case for using continuum-scored words and word shift graphs. EPJ Data Science, 28, 6, 2017.
[pdf] [arXiv]
Peter Sheridan Dodds, David Dewhurst, Fletcher F. Hazlehurst, Colin van Oort, Lewis Mitchell, Andy Reagan, Jake Williams, Chris Danforth. Simon's fundamental rich-gets-richer model entails a dominant first-mover advantage. Physical Review E, , 95, 2017.
[pdf] [journal page] [arXiv]
Sharon Alajajian, Jake Williams, Andy Reagan, Stephen C. Alajajian, Morgan Frank, Lewis Mitchell, Jacob Lahne, Chris Danforth, Peter Sheridan Dodds. The Lexicocalorimeter: Gauging public health through caloric input and output on social media. PLoS ONE, e0168893, 12, 2017.
[pdf] [journal page] [arXiv]
Eric M. Clark, Jake Williams, Chris Jones, Richard Galbraith, Chris Danforth, Peter Sheridan Dodds. Sifting robotic from organic text: A natural language approach for detecting automation on Twitter. Journal of Computational Science, 1-7, 16, 2016.
[pdf] [journal page]
Eric M. Clark, Chris Jones, Jake Williams, Allison Kurti, Mitchell Norotsky, Chris Danforth, Peter Sheridan Dodds. Vaporous marketing: Uncovering pervasive electronic cigarette advertisements on Twitter. PLoS ONE, , 11, 2016.
[pdf] [journal page]
Jake Williams, Jim Bagrow, Andy Reagan, Sharon Alajajian, Chris Danforth, Peter Sheridan Dodds. Selection models of language production support informed text partitioning: an intuitive and practical bag-of-phrases framework for text analysis. Preprint, 2016.
[pdf] [arXiv]
Jake Williams, Jim Bagrow, Andy Reagan, Sharon Alajajian, Chris Danforth, Peter Sheridan Dodds. Zipf's law is a consequence of coherent language production. Preprint, 2016.
[pdf] [arXiv]
Andy Reagan, Brian Tivnan, Jake Williams, Chris Danforth, Peter Sheridan Dodds. Benchmarking sentiment analysis methods for large-scale texts: A case for using continuum-scored words and word shift graphs. Preprint, 2015.
[pdf] [arXiv]
Jake Williams, Eric Clark, Jim Bagrow, Chris Danforth, Peter Sheridan Dodds. Identifying missing dictionary entries with frequency-conserving context models. Physical Review E, 042808, 92, 2015.
[pdf] [journal page] [arXiv]
Jake Williams, Peter Sheridan Dodds, Chris Danforth, Jim Bagrow, Eric Clark, Suma Desu, Paul Lessard. Zipf's Law holds for phrases, not words. Nature Scientific Reports, 12209, 5, 2015.
[pdf] [journal page] [arXiv] [online appendices]
(show all)

Most recent press: