I am working with Dr. Judith Gelernter on Geo-Information Extraction on Twitter data. The project is on google code. This is a machine-learning based and rule-based named entity recognizer for twitter. It could handle English and Spanish tweets. We used over 3000 location-tagged tweets for Spanish, about 6000 English location-tagged tweets to train linear-chain CRF and Voted Perceptron Hidden Markov Model(VPHMM).
-Java, Python, matlab
-Fluent English, a little bit of Russia
I am a huge music fan! I was raised in the northeastern China, so heavy snow in Pittsburgh is perfectly fine with me! After graduation from HIT, I moved to Beijing, and lived there for 5 years until I came to Pittsburgh. I love making new friends, cooking, and playing piano.
BTW, I am gonna be a DAD soon! A baby boy's coming!
Find me on LinkedIn Facebook Twitter Renren Weibo