MSDSA Project Descriptions (2024 Cohort)
Masters project descriptions for the 2024 cohort of Masters in Data Science and Analytics students. Final project defense slides will be added upon completion.
Pushpak Rane
Integrating Indian Sign Language (ISL) into Healthcare for Deaf Patients in India.
In India, over 18 million deaf and hard-of-hearing patients face barriers to effective healthcare communication. Although Indian Sign Language (ISL) is officially recognized, it is rarely implemented in medical settings due to a shortage of trained professionals, leading to miscommunication and inadequate treatment. This project develops an AI-based system for real-time ISL interpretation using computer vision, trained on medical sign datasets, to bridge communication gaps and improve healthcare accessibility for the Deaf community.
Khushi Choudhary
Optimizing Volunteer Scheduling and Emergency Response Using Machine Learning
This project aims to build a predictive, real-time volunteer scheduling system that connects operational needs with available resources. Currently, incident command struggles with limited visibility—uncertain who will arrive, when, and with what skills—while volunteers lack clarity on where they are needed, particularly across mutual aid groups. With three separate scheduling systems in play (NVADG, County, and Mutual Aid), coordination is chaotic. The new app will apply machine learning to forecast shortages and skill gaps, provide volunteers with clear shift visibility, and enable coordinators to make faster, data-driven decisions in real time.
Govardhan Baddala
Offline-First Web App for Collecting and Syncing Geology Field Data
This project develops an offline-first web app for geology fieldwork, designed for use in remote areas without internet access. Currently, students rely on paper notes, which can be lost or introduce errors during later entry. The app stores data locally, then automatically syncs once a connection is available, ensuring faster, more accurate collection and reducing the risk of lost observations.
Shivam Pawar
GraphLearnR: An R Package for Powerful and Accessible Graph Learning
Graph learning is an emerging field that uncovers hidden network structures from observed data, with applications ranging from social networks and brain connectivity to protein interactions and transportation systems. Current methods often assume signal smoothness, overlook external factors, and require advanced coding skills. This project addresses these gaps by developing regression-aware graph learning tools in R, making advanced network structure discovery more accessible to researchers in neuroscience, social science, and biology.
Nate Dailey
Raster-Based Wildfire Risk Model in Python
This project develops a scriptable wildfire model in Python to help identify hazardous areas for treatment and reduce wildfire risk. By calculating rate of spread, fireline intensity, and flame length for each pixel, the model provides a technical basis for prioritizing treatments such as mechanical thinning or prescribed burns. Inputs include fuel model and slope rasters, as well as wind speed, wind direction, and fuel moisture, enabling flexible, data-driven fire behavior analysis.
Zakir Elaskar
B-Line Public Transit: A Data-Driven Analysis for Service Optimization This project applies data science techniques to analyze B-Line, the public transportation system serving Chico and surrounding communities. Public transit is essential for students, workers, and residents, yet inefficiencies can limit service quality and cost effectiveness. By examining trends, detecting inefficiencies, and proposing optimization strategies, this work provides data-driven insights to improve service delivery and promote more sustainable, cost-effective urban mobility.
Snehitha Gorantla
Air Pollution Prediction using Time Series Analysis
This project develops a hybrid time series model to forecast air pollutant levels in California using EPA CASTNET data (1987–2025). By incorporating meteorological variables such as temperature, wind, and humidity, the model aims to improve the accuracy of short-term air quality predictions. The research explores how weather factors influence pollutant concentrations over time and evaluates the extent to which these variables enhance predictive precision across both rural and suburban regions.
Abinesh S
AI Driven Medical Case Manager
This project develops an AI-driven medical case manager designed to address the nationwide shortage of trained case managers in healthcare. The system replicates key human functions—including care coordination, advocacy, resource navigation, and health monitoring—while focusing on older adults (65+) with complex needs. By integrating patient advocacy and continuous monitoring, the AI aims to reduce fragmented care, minimize delays in treatment, and improve outcomes. Research will assess its impact on hospital readmissions, patient satisfaction, and workflow integration, while also exploring feasibility within existing healthcare infrastructures.
Amol Bhalerao
Leveraging Machine Learning to Model CalFresh Participation CalFresh, California’s version of SNAP, provides essential food assistance to low-income individuals, including college students. Yet, despite 40% of California’s college students being eligible, 82% do not access these benefits. This gap is critical, as 1 in 3 students experience food insecurity, which can cause stress, poor nutrition, and difficulty focusing on academics. Using data from CHC-UCLA, this project will use machine learning to examine the barriers that prevent students from enrolling in CalFresh and identify strategies to increase participation, with the goal of improving student well-being and academic success.
Anand Gangavarapu
Evaluating the Ecological Effects of Mastication and Wildfire in Chaparral
In summer 2021, the Park Fire burned through parts of the Big Chico Creek Ecological Reserve (BCCER) in Butte County, California, impacting chaparral ecosystems and previously treated management units. While mastication is a widely used fuel treatment in chaparral, its ecological outcomes after wildfire remain uncertain. This project evaluates how masticated chaparral responds to fire, providing insight into treatment effectiveness, ecosystem recovery, and long-term resilience.
Jayana Sarma
Morphometric Sex Determination and GPS Telemetry of Turkey Vultures in Western Montana
Turkey Vultures (Cathartes aura) play a vital ecological role as scavengers, yet key aspects of their biology in Western Montana remain poorly understood. This study focuses on the subspecies C. a. meridionalis, which exhibits both migratory and resident behaviors. Research goals include developing accurate morphometric models to determine sex—addressing the challenge of visually similar males and females—while correcting field data inconsistencies. In addition, GPS telemetry is used to track movements and locate hidden nesting sites, with interactive visualizations created to clearly communicate findings. Together, these methods provide new insights into vulture ecology, migration, and breeding in Montana.