Julian Brooke

I was a PhD student and postdoctoral fellow in the Department of Computer Science at the University of Toronto (Now at the University of Melbourne).

Contact

Interests

My general area of interest within CS is computational linguistics (CL). The research I do is focused mainly on semantics and style, in particular the stylistic aspects of the lexicon, sentiment analysis, and native language identification. I've also been employed to work on projects in other areas of CL, including syntatic and discourse parsing, word prediction, and entity recognition. In my work I use a fairly wide range of statistical and rule-based approaches to CL problems, but I have a certain preference for "big data" methods, i.e. unsupervised and semi-supervised approaches that leverage large internet corpora. I enjoy cross-disciplinary research in general, including applications in education, linguistics, and analysis of literature.

Outside of school, I like travelling and reading. I've studied various languages, and I speak fluent Mandarin Chinese, thanks to several years living in China and the ongoing patience of my wife.

Papers, posters, and presentations

• Brooke, Julian and Graeme Hirst. 2014. Unsupervised multiword segmentation of large corpora using prediction-driven decomposition of n-grams. In Proceedings of COLING 2014. Presentation and paper.
• Brooke, Julian and Graeme Hirst. 2014. Supervised ranking of co-occurrence profiles for acquisition of continuous lexical attributes. In Proceedings of COLING 2014. Presentation and paper.
• Tagliamonte, Sali A. and Julian Brooke. 2014. A weird (language) tale: Varation and change in adjectives of strangeness. American Dialect. Journal Paper.
• Brooke, Julian and Graeme Hirst. 2013. Hybrid models for lexical acquisition of correlated styles. In proceedings of IJCNLP 2013. Presentation and paper.
• Brooke, Julian and Graeme Hirst. 2013. Investigating the effect of multi-L1 learner corpora variables on native language identification. Presented at LCR 2013.
• Brooke, Julian and Graeme Hirst. 2013. A multi-dimensional Bayesian approach to lexical style. In Proceedings of NAACL 2013. Presentation and short paper.
Brooke, Julian, and Graeme Hirst. 2013. Using external corpora in the 2012 NLI shared task. In Proceedings of BEA-8 Workshop, NAACL 2013 . Poster and paper.
• Brooke, Julian, Graeme Hirst, and Adam Hammond. 2013. Clustering voices in The Waste Land. In Proceedings of the CLFL Workshop, NAACL 2013. Poster and short paper.
• Hammond, Adam, Julian Brooke, and Graeme Hirst. 2013. A tale of two cultures: Bringing literary analysis and computational linguistics together. In Proceedings of the CLFL Workshop, NAACL 2013. Presentation and paper.
• Brooke, Julian and Graeme Hirst. 2013. Multidimensional analysis versus latent semantic analysis for constructing a register space: Are hand-coded features needed or is bags-of-words enough? Presented at LSB 2013. Presentation.
• Brooke, Julian and Graeme Hirst. 2012. Robust, lexicalized native language identification. In Proceedings of COLING 2012. Presentation and paper.
• Brooke, Julian and Graeme Hirst. 2012. Paragraph clustering using a stylistic vector space model with extrinsic features. In Notebooks of the PAN 2012 Shared Task, CLEF 2012. Poster and paper.
• Brooke, Julian, Adam Hammond, and Graeme Hirst. 2012. Unsupervised stylistic segmentation of poetry with change curves and extrinsic features. In Proceedings of the CLFL Workshop, NAACL 2012. Presentation and paper.
• Brooke, Julian, Vivian Tsang, David Jacob, Frasier Shein, and Graeme Hirst. 2012. Building readability lexicons with unannotated corpora. In the Proceedings of the PITR Workshop, NAACL 2012. Presentation, poster, and paper.
• Brooke, Julian and Graeme Hirst. 2012. Measuring interlanguage: Native language identification with L1-influence metrics. In the Proceedings of LREC 2012. Presentation and paper.
• Brooke, Julian and Graeme Hirst. 2012. Facets of formality: A dimension of register in a sociolinguistic corpus. Presented at GURT 2012. Poster.
• Brooke, Julian and Sali Tagliamonte. 2012. Hunting the linguistic variable: Using computational techniques for data exploration and analysis. Presented at GURT 2012. Presentation.
• Brooke, Julian, Tong Wang, and Graeme Hirst. 2011. Predicting word clipping with latent semantic analysis. In the Proceedings of IJCNLP 2011. Presentation and short paper.
• Brooke, Julian and Graeme Hirst. 2011. Native language detection with ‘cheap’ learner corpora. Presented at LCR 2011. Presentation and paper.
• Brooke, Julian and Graeme Hirst. 2011. Lexicalizing computational stylistics for language learner feedback. In the Proceedings of SAD 2011. Presentation and paper.
• Taboada, Maite, Julian Brooke, Milan Tofiloski, Kimberly Voll, and Mandred Stede. 2011. Lexicon-based methods for sentiment analysis. Computational Linguistics: 17 (2). Journal paper.
• Brooke, Julian, Tong Wang, and Graeme Hirst. 2010. Automatic acquisition of lexical formality. In the Proceedings of COLING 2010. Poster and paper.
• Brooke, Julian. 2010. Computational Approaches to Normative Style. Qualifying Oral Examination, Univerisity of Toronto. Paper.
• Brooke, Julian, Tong Wang, and Graeme Hirst. 2010. Inducing lexicons of formality from corpora. In Proceedings of the Workshop on Automatic Acquisition of Language Resources, LREC 2010. Poster and paper.
• Brooke, Julian and Matthew Hurst. 2009. Patterns in the stream: Exploring the interaction of polarity, topic, and discourse in a large opinion corpus. In Proceedings of the Workshop on Topic-Sentiment Analysis, CIKM 2009. Presentation and paper.
• Brooke, Julian, Milan Tofiloski, and Maite Taboada. 2009. Cross-Linguistic sentiment analysis: From English to Spanish. In Proceedings of RANLP 2009. Poster and paper.
• Taboada, Maite, Julian Brooke, and Manifred Stede. 2009. Genre-based paragraph classification for sentiment analysis. In Proceedings of SIGDIAL 2009. Paper.
• Tofiloski, Milan, Julian Brooke, and Maite Taboada. 2009. A Syntactic and lexical-based discourse segmenter. In Proceedings of ACL-IJCNLP 2009. Short paper.
• Brooke, Julian. 2009. A Semantic Approach to Text Sentiment Analysis. Unpublished MA thesis. Simon Fraser University.
• Brooke, Julian. 2009. The quantification of sentiment. Presented at the WISSLR Symposium. Presentation.
• Taboada, Maite, Kimberly Voll, and Julian Brooke. 2008. Extracting Sentiment as a Function of Discourse Structure and Topicality. Simon Fraser University School of Computing Science Technical Report 2008-20.
• Brooke, Julian, Greg Coppola, Vivian Lee, and Yang Zhao. 2008. Phrasing, tempo, and focus: A Prosodic-Hierarchy approach to Mandarin tone sandhi domains. In Proceedings of the ICEAL2 . Presentation and paper.
• Brooke, Julian. 2008. Light NP shift and verbal adjuncts in English. In Proceedings of the CLA Conference 2008. Presentation and paper.

Resources

• List of formal and informal words from my work on formality.
• A list of URLs and a web scrapper for recreating the Lang8 non-native English corpus used in my NLI work. Note the web scrapper might need to be updated (as the website changes), and a name and password on Lang8 are required to extract the full corpus.
• Various annotations of voice switches in the poem The Waste Land , used in my unsupervised stylistic segmentaiton work.
• A set of 5 annotations for 900 words of varying style, used in my work on correlated styles (the 6 style model).
• The n-gram decomposition segmenter tool, from my work on multiword segmentation
• Raw polarity/SO scores derived from the ICWSM corpus using the methods in my paper on supervised ranking of co-occurence profiles.
• The LPR n-gram decomposition formulaic sequence identifier, basically an upgraded version of the n-gram decomposition segmenter, plus related materials (a lexicon and test sets).
• The GutenTag tool for digital humanities studies in the Project Gutenberg corpus.