The NLP Research Project
The purpose of our project is to conduct NLP research and present our findings in the form of academic papers. This entails multifaceted approaches to the problem, including managing extremely large datasets, building complex machine learning models, performing statistical validation, and analyzing the outcomes.
We believe that the student-run research allows us to apply our knowledge and learn the materials beyond what courses can offer. We constantly explore cutting-edge technology in machine learning and deep learning at the academic research level: this academic incentive is something that drives us forward. Weekly presentations facilitate our members to teach and learn the forefront of machine learning research.
This past semester, we had a wide range of research topics, from recommendation system to deep style transfer. In general, we took the approach called Natural Language Processing -- an interaction between machine learning and text analysis. Our student groups studied hidden topics in texts to make a recommendation, analyze trends, or vectorized texts to generate new reviews.
With the help of our faculty advisor, Thorsten Joachims, our team has submitted 4 papers to the Yelp Dataset Challenge -- A Data Science research competition hosted by Yelp. All research projects demonstrated remarkable results: an implementation of recommendation system that beats industry standard algorithms, an accurate analytic tool to assess “trends” of business, a classifier to identify locally popular users, and a tool that transfers writing style using deep learning.
To learn more, visit our GitHub here.