Tokenizer and Beyond

Wrote the tokenizer last week. Works nicely.

Started the new and improved (i.e. GPU version) word count functions this weekend.

Figured out a new and better question to ask. Rather than asking the political sway of an article - which is too hard to even ask humans to label - my tool will try to determine the source of an article (CNN, FOX, etc), which is sort of the same question, but easier to annotate.

Started thinking of how to parallelize the learning algorithm.

old beginnings, new beginnings

Wrote a word count program on the cpu this weekend. Turned out to be not what the doctors ordered....

Started over last night. Wrote the beginnings of a tokenizer on the gpu. My emulator at home was giving me problems, spent a while debugging.

Hoping to finish the tokenizer + get a tagged corpus this week.

Hello World!

Well here I go with my first blog...

This blog is being set up for my CIS665 course at the University of Pennsylvania, where I am currently a 3rd PhD student in the Center for Human Modeling and Simulation.

As required by the class, this blog will track the trials and tribulations of my project, titled "Are you an Obama girl? A GPGPU classification of semantic orientation in political articles." A complete explanation of the thesis of this project will be posted soon.