Parikshit Solunke

MOUNTAINEER: Topology-Driven Visual Analytics for Comparing Local Explanations

Parikshit Solunke, Vitoria Guardieiro, Joao Rulff, Peter Xenopoulos, Gromit Yeuk-Yin Chan, Brian Barr, Luis Gustavo Nonato, Claudio Silva

HuBar: A Visual Analytics Tool to Explore Human Behaviour based on fNIRS in AR guidance systems

Sonia Castelo, Joao Rulff, Parikshit Solunke, Erin McGowan, Guande Wu, Iran Roman, Roque Lopez, Bea Steers, Qi Sun, Juan Bello, Bradley Feest, Michael Middleton, Ryan Mckendrick, Claudio Silva

Granularity at Scale: Estimating Neighborhood Socioeconomic Indicators from High-Resolution Orthographic Imagery and Hybrid Learning

Ethan Brewer, Giovani Valdrighi, Parikshit Solunke, Joao Rulff, Yurii Piadyk, Zhonghui Lv, Jorge Poco, Claudio Silva

Mountaineer: Topology-Driven Visual Analytics Tool

Developed "Mountaineer" in collaboration with Capital One; A Topology-Driven Visual Analytics tool designed to facilitate the comparison of Machine Learning Explanations. Published in IEEE Transactions on Visualization and Computer Graphics.

• Conceptualized and designed the user interface.
• Implemented and developed the tool, applying topological concepts and visualization to compare ML Model Explanations.
• Designed case studies and conducted interviews with industry experts to evaluate the tool.

HuBar: Visual Analytics for Human Behavior Analysis

Contributed to HuBar, a tool that integrates cognitive workload and multimodal sensor data for analyzing and comparing performer behavior during AR assisted-task performance, as part of DARPA's Perceptually-enabled Task Guidance (PTG) Project. Published at IEEE VIS 2024.

• Designed and developed the visual interface for HuBar.
• Assisted in gathering expert feedback through interviews.

OpenSpace: Apple Silicon Port

Developed an Apple silicon compatible port for OpenSpace, a data visualization software to visualize the entire known universe. This expanded compatibility creates new research opportunities, including developing hybrid natural language and VR-based interaction methods, particularly for the Apple Vision Pro.

GDPFinder: Estimating Neighborhood Well-Being

Developed GDPFinder, a project which focuses on estimating neighborhood well-being from high-resolution satellite imagery using supervised and semi-supervised learning techniques. Published in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

• Fine-tuned a ResNet-50 based architecture within the supervised methodology to predict GDP, educational attainment, and population density at a block level in US cities.
• Designed a visualization tool to interpret and analyze auto-encoder and clustering results in the semi-supervised methodology.

EuroVis 2021 Committee Network Visualization

A visualization documenting academic relationships between committee members of EuroVis 2021- An annual dataviz conference. Made as a part of my Master's Degree project. Frontend built using HTML, CSS, JavaScript and D3.js. Majority of the academic connections were scraped using the Google Scholar API, and data was cleaned Using pandas.

Employment Precarity Visualization

A visualization aiming to showcase the deteriorating quality of employment in the US in the recent past. Made with HTML, CSS, Bootstrap, JavaScript, and D3.js.

A Day at the Movies

A visualization of the Internet Movie Database (IMDB). Made with R and Shiny.

Selective Detective

A VR game developed in Unity(C#) for the state of the art CAVE2™ system, a large-scale virtual-reality environment at UIC's Electronic Visualization Laboratory.

New York University

Visualization Imaging and Data Analysis Center

Advised by Dr. Claudio Silva

University of Illinois at Chicago

GPA: 3.87