I developed a web application that uses an emotion classifier to classify a user's mood and analyzes the data over time. The website is hosted at https://mood-tracker-sg.herokuapp.com/.
My team and I, under the guidance of Professor Kilian Weinberger, created Anonymization Station. Anonymization Station is a tool that detects and anonymizes faces in an image and reverses the anonymization when provided the appropriate key. Below is a demo of the anonymization website.
My partner and I developed a one-player 3D version of Breakout. The paddle can be controlled using the user's cursor, and the ball bounces off the paddle at an angle proportional to its distance from the center of the paddle.
WIP
For our final project, my partner and I built an interpreter from scratch for a custom object-oriented language similar to Java. Our intention was for the language to be less verbose and complex, but still support important features such as class inheritance. Below is a demo of our interactive interpreter.
For our midterm project, my team and I developed an OCaml chatbot that parses through a corpus of 100+ Wikipedia articles to answer questions related to the field of computer science. We utilized NLP/IR algorithms such as TF-IDF, Jaccard, and Cosine Similarity to identify the most probable answer & provide suggestions to typos in user queries.