Students in the course, which aims to ground the data-driven techniques used in language technologies in sound statistical methodology, must complete a substantial team project at the end of the semester. The project requires students to discover and exploit deficiencies in the conventional trigram language model using sound statistical methods by writing a program that tries to distinguish real broadcast news articles from fake ones generated by a broadcast-news-trained trigram model. The performance of each team's program is judged on both a hard metric — the percent of items correctly classified; and a soft metric — the average log posterior of the correct label. Lian and Liu achieved the best results by both metrics.
MLT student David Klaper joined Lian and Liu on the leaderboard as the top student contributor for participation.
The winners will receive chocolate truffle prizes at the LTI's graduation celebration on Friday, May 15.
Read more about the class on the course home page.