Machine Learning: Where Data-Driven Student Projects are Born

First and second-year computer science students at NYU Shanghai impressed with data-driven projects from Machine Learning with Professor Keith Ross and Object Oriented Programming with Professor Promethee Spathis, using artificial intelligence and data science in creative, practical, and innovative applications.

Tired of reading long emails? Fortunately, Kenny Song and Quan Vuong created “uMail,” a working system using artificial intelligence to automatically summarize and highlight the most important details from emails as a Gmail-integrated Chrome extension.

As one of the most difficult aspects of learning the Chinese language is perfecting its tones, Jennifer Huang and Kelvin Liu devised “Mandarin Tone Learning,” a working app that uses machine learning to help Mandarin learners improve their tone pronunciation.

Ambitiously, Cameron Ballard and Carson Nemelka explored whether computers can replace doctors with “Predicting Diabetic Retinopathy,” an application that predicts whether a patient has a common diabetic eye disease—a leading cause in adult blindness—based solely on images of the eye.

Jessica Chen examined the possibilities of social networks predicting stock market performance with “Quantifying Stock Volatility Using Sentiment Analysis.” She found that Sentimental Analysis on tweets has been proven with certain accuracy to predict the index volatility and aims to expand the analysis to discussion text and analytical reports of specific stocks.

Another data-driven project from the Object Oriented Programming course is “SeeMetro” by Watcher Wu and Autumn Wu. It started with raw train data made publicly available by the NYC MTA and developed into an insightful real-time visualization of all MTA subways—user-friendly and keyboard interactive to explore the ever-moving and colorful mapping of train networks.

Watch the project videos on Youtube

Watch project videos on Youku

Written by Charlotte San Juan