Lecturer · PhD student · NJIT
I teach data visualization and programming at the New Jersey Institute of Technology, and I research how visual analytics can make machine learning models easier to understand.
I am a full-time lecturer in the Department of Data Science at NJIT and a part-time PhD student at the same university. I earned my BS at NJIT, did research here as an undergraduate, and joined the faculty as a lecturer. Counting my time as a student, I have been part of NJIT for about ten years.
My work sits where teaching and research meet. In class I help students learn to turn data into clear visuals and working code. In research I look for ways to let people see what a model is doing and why.
My research and publications focus on data visualization and on interpretable AI through visual analytics. I also apply machine learning to financial data.
Designing views that help people read patterns, spot problems and make decisions from complex data.
Using visual analytics to explain what a model learned and where it can go wrong.
Applying machine learning to financial data and keeping models reliable as markets change.
I am currently working on distribution shift in financial data: what happens to a model when the data it sees after deployment no longer looks like the data it was trained on. Markets, customers and regulations change, so models built on financial data can quietly lose accuracy. I study how to detect these shifts and how to make models hold up under them.
How to design, build and critique visualizations that communicate data clearly and effectively.
The principles and practice of turning data into clear, effective visualizations.
A first course in programming in Python for students across majors.
J Yuan, K Bhattacharjee, AZ Islam, A Dasgupta
The Visual Computer 40 (5), 3615–3631
K Bhattacharjee, A Islam, J Vaidya, A Dasgupta
2022 IEEE Symposium on Visualization for Cyber Security (VizSec), 1–11
K Bhattacharjee, A Islam, A Dasgupta
A Islam, A Dasgupta