Current Students


Atharv Yeolekar

MS Student [2021 - 2023]

Data Science and Analytics

TBD


[Work in progress.]

TBD


Krishna R. Puthucode

MS Student [2021 - 2023]

Computer Science

"TBD"


[Work in progress.]

TBD


Shreejaa Talla

MS Student [2021 - 2023]

Computer Science

"Augmentation of Solar Filaments for Machine Learning"


[Work in progress.]

The magnetic structure of the Sun is evident in many large-scale events of the Sun, such as solar filaments. Sudden disappearance of filaments is well associated with coronal mass ejections which cause threats to our lives on Earth. One of the most informative characteristics of filaments is their chirality that can be determined by their shape and texture. While this is a perfect image classification task, manually identifying filaments' chirality is a tedious task that requires domain experts. Having a relatively small dataset of manually labeled filaments at our disposal, Shreejaa is currently developing algorithms to augment filaments by applying a series of transformation function on them. One of the main scientific challenges of this task is to generate new instances in such a way that no underlying physical theories are broken in the augmented instances. The goal is to put this augmentation engine in the form of an Image Loader similar to torch.utils.data.DataLoader.

Former Students


Egill Gunnarsson

MS Student [2019 - 2021]

Computer Science

"Interactive Supervised Machine Learning Model Evaluation using D3"


[Defense][Final Product]

The evaluation of a supervised machine learning model is one of the most important aspects of its life cycle. While there are numerous evaluation metrics each of which provides a different insight into models’ performance it can be sometimes challenging to find the appropriate ones that fit the problem in hand. Including the imbalance ratio as an extra variable makes the evaluation process even more difficult. Therefore, I implemented a web application to intuitively evaluate models' performance based on their confusion matrices and given imbalance ratios. This project is an online, interactive application of the Contingency Space recently proposed by Ahmadzadeh et al. (2021). Inspired by this concept, my web application allows the user to visually evaluate their pre-trained supervised models. A side-by-side graphical representation of multiple metrics is provided for a comparison between each metrics score. Confusion matrices are evaluated on metrics such as Accuracy, Precision, F1-Score, Recall, etc. Additionally, the user can load their own customized metrics as well. The visualization is based on contour plots that correlate to each metrics score in relation to True Positive and True Negative rates and imbalance ratios.

This application uses technologies such as d3.js, Python, JavaScript, html, css, flask and json. Each metric’s score is generated in the backend using Python, based on an imbalance ratio. Information is sent to and from the backend via flask and json objects. JavaScript then uses the d3 library to convert the metric scores into a contour plot. The d3 library has many interactive capabilities which allows the user to modify the evaluation to fit every requirement.


Sonam Dawani

MS Student [2019 - 2021]

Data Science and Analytics

"A Texture-Based Approach for Identification of Filaments’ Chirality" [video]

S. Dawani, A. Ahmadzadeh, and R. A. Angryk

43rd COSPAR Scientific Assembly (2021) [+], Machine Learning for Space Science Workshop (ML4SS) [+][+][programme]

Kankana Sinha

MS Student [2017 - 2019]

Computer Science

"MVTS-Data Toolkit: A Python Package for Preprocessing Multivariate Time Series Data" [pdf]

Azim Ahmadzadeh, Kankana Sinha, Berkay Aydin, and Rafal A. Angryk

SoftwareX [+] Journal (Elsevier), 2020