Congratulations to Adrija Ghosh, the winner of the 2026 Friends of the Library Scholarship! Adrija is pursuing a Master’s degree in Data Science, and is set to graduate in the fall of 2026.
Adrija was chosen for the scholarship based on her essay response to the following prompt:
How did you use library resources and services to complete a project, and what did they add to your work?
During the development of my course project, “STORM-CARE: Uncertainty-Aware Hurricane Forecasting and Child-Centered Decision Support,” for COMP SCI 5542 – Big Data Analytics & Applications (Spring 2026), I extensively utilized the University of Missouri–Kansas City (UMKC) Miller Nichols Library resources to strengthen both the technical and research foundations of my work.
The project required integrating advanced concepts in spatio-temporal modeling, geospatial analytics, and humanitarian decision-support systems. To build a strong conceptual framework, I relied on the UMKC library’s access to scholarly databases such as IEEE Xplore, ACM Digital Library, and ScienceDirect. These platforms enabled me to review peer-reviewed research on graph neural networks (GNNs), probabilistic forecasting, and climate data modeling. In particular, literature on spatio-temporal graph learning and uncertainty quantification helped guide the design of my forecasting model and evaluation framework.
Additionally, the library’s access to interdisciplinary journals was crucial in connecting technical machine learning approaches with real-world humanitarian applications. I explored publications related to disaster management, social vulnerability, and public health analytics, which informed the child-centered risk indicators used in my project. This helped ensure that my system design was not only technically sound but also socially meaningful and aligned with humanitarian planning workflows.
The UMKC library’s research guides and digital resources also supported my work in sourcing and validating datasets. Through library-recommended sources and data repositories, I identified authoritative datasets such as NOAA’s HURDAT2 hurricane database, ERA5 climate reanalysis data, and social vulnerability indices. The credibility and traceability of these data sources were essential for ensuring reproducibility and transparency in my project.
Furthermore, citation management tools and academic writing support provided through the library enabled me to organize references efficiently and maintain proper academic integrity. These tools were especially helpful when compiling the research white paper, ensuring correct attribution and consistent formatting of sources.
The library environment itself—both physical and digital—played an important role in supporting focused research and productivity. Access to quiet study spaces and reliable digital infrastructure allowed me to work on complex tasks such as data preprocessing, model development, and visualization design without interruption.
Overall, the UMKC library resources significantly enhanced the quality and depth of my project. They enabled me to bridge theoretical knowledge with practical implementation, incorporate credible and diverse data sources, and produce a research-driven, reproducible system. The library’s contribution extended beyond information access it supported critical thinking, interdisciplinary integration, and the development of a meaningful, real-world application. This experience highlights the essential role of academic libraries in advancing student research and innovation.
