User:Kwmurray

I'm a second year Master's Student in LTI. I'm interested in unsupervised Machine Learning techniques, and in particular, graphical models. I'm hoping to work on some graphical model project for the final and would be interested in doing something with Weibo - but do not speak any Chinese - so if anyone who does is looking for a project partner, I might be interested.

In general, what I'm hoping to get out of the class is a better understanding of how to deal with massive amounts of noisy free text data. I'm fascinated by how language changes when people are using social media and think it is an interesting robustness check for any model. Normal emphases on spelling and grammar when writing in a formal context are often ignored in social media, but the semantic meanings of a text are still conveyed through this noisy channel.

Wiki Write-up Assignments (Social Media Analysis 2012)

• Papers

1. Jonathan Chang, Jordan Boyd-Graber, David Blei. 2009. Connections between the Lines: Augmenting Social Networks with Text. In KDD 2009
2. Seth Myers, Chenguang Zhu, Jure Leskovec. 2012. Information Diffusion and External Influence in Networks. In KDD 2012
3. Jacob Eisenstein, Amr Ahmed, and Eric Xing. 2011. Sparse Additive Generative Models of Text. In ICML 2011

Project (Social Media Analysis 2012)

• Proposal