profile pic

Sejal D

- Research Program Mentor

MS Master of Science candidate


Data Science, Data Analytics, Machine Learning, Sports Analytics, Natural Language Processing, Data Visualization, Computer Science, Biotechnology

Project ideas

Project ideas are meant to help inspire student thinking about their own project. Students are in the driver seat of their research and are free to use any or none of the ideas shared by their mentors.

Sentiment analysis of COVID-19 vaccine tweets

Apply data mining to query and synthesize hundreds of thousands of tweets and perform sentiment analysis to compare which COVID-19 vaccine is most promising in different geographic regions. Compare which side effects are most predominant among Pfizer vaccine recipients versus Moderna vaccine recipients. This project would likely culminate in a Medium article which takes the reader through the project from exploratory data analysis to code implementation, and finally a well-articulated discussion of research findings and limitations.

Using biomedical sensing data to examine indicators of cognitive decline over time

Providing that there is accessible sensing data collected from Dementia and Alzheimer's patients, analysis can be performed on gait (walking speed), balance, and circadian rhythm changes over time. Perhaps the student can train a classifier to predict the likelihood that an undiagnosed elderly person has cognitive impairment given their health data over some extended period of time. This project could culminate in an interactive web app or a research paper / blog post.

Neuroimaging Classification: a machine learning approach for glioma detection

Given brain MRI scans, the student will apply image processing techniques to get usable black-and-white image image objects and feed them into a classifier to automatically extract tumor information from the scans, thus reducing the burden on imaging specialists by augmenting the task of medical diagnosis with AI and technology.

NBA shot selection analysis using SportVU tracking data

The NBA releases an abundance of coordinate-based tracking data for each game. This project will make use of the 25 frames of data per second to not only identify when shots have been taken but also retrieve key information about the selection of shots by each team and player. The student will build an app to display shot charts and also potentially classify how smart a shot is based on metrics like score differential, shot distance, defender distance, shot clock usage, etc. This will help build a compelling analysis of what differentiates good shooters from smart shooters!

Coding skills

Python, Javascript (React), SQL, C, C++, Matlab, HTML, CSS


"Sejal is a really great mentor and I'm so happy to have been paired with her! She is so dedicated and flexible and she is very helpful both technically and just for guidance. I can't believe I wrote my first research paper and it was a great experience because of her!"

Pavithra from Dublin, CA

Pavithra from Dublin, CA profile image

Interested in working with expert mentors like Sejal?

Apply now