Current and past projects and publications.
My Portfolio
Current Projects
NABAT: North America Bat Monitoring program
NABat is a multi-national, multi-agency coordinated bat monitoring program across North America. This collaborative bat monitoring program is made up of an extensive community of partners across the continent who use standardized protocols to gather data that allow us to assess population status and trends, inform responses to stressors, and sustain viable populations. I am currently working alongside other developers to maintain and implement new features that support ecologists in their ongoing bat conservation efforts.
Fort Collins Science CEnter authentication
The developers within the Ecotech branch in the Fort Collins Science Center maintains a suite of apps that require user accounts for managing uploads, editing sites, and tracking data. These logins support both internal and external partners through Microsoft Entra ID and Login.gov integration. I deployed the Keycloak application to a virtual private cloud using Amazon's Cloud Development Kit, alongside a CI/CD pipeline managed by GitLab runners. Additionally, I develop and maintain the authentication system, along with a suite of internal packages for Angular and R, to ensure seamless integration with the Fort authentication service.
Sagestep
I am currently working on the Sagebrush Steppe Treatment Evaluation Project, a regional study focused on evaluating methods for restoring sagebrush steppe ecosystems in the Great Basin. In collaboration with scientists, I’m developing a new site using R and Posit to authenticate and access project data. This new platform aims to provide ecologists with greater control and visibility over their data. By transitioning the site and calculations into R, we make the platform more accessible and interpretable for sagestep scientists, enhancing their ability to analyze and interpret results.
HDGov is a clearinghouse for social science information related to interactions between people and the environment. The website features job postings, training opportunities, and publications. It was developed using TypeScript, with a tech stack that includes MongoDB, Node.js, Express.js, and Angular.
Human Dimensions (HDGOv)
WNS is a clearinghouse and information center dedicated to tracking the spread and control of White-Nose Syndrome. The website features an interactive map built with ArcGIS, along with a reporting portal that allows ecologists to upload their test results on bats affected by the disease. It was developed using MongoDB, PostgreSQL, Node.js, Express.js, and Angular.
White Nose Syndrome(WNS)
Past Projects
Actev: Activities in extended video
ActEV is a series of evaluations to accelerate the development of robust, multi-camera, automatic activity detection algorithms for forensic and real-time alerting applications. ActEV is an extension of the annual TRECVID Surveillance Event Detection (SED) evaluation where systems will also detect and track objects involved in the activities. Each evaluation will challenge systems with new data, system requirements, and/or new activities. Currently we are running the ActEV 2021 Sequestered Data Leaderboard (SDL) evaluation that features Unknown Facility and Surprise Activity Testing and the ActEV TRECVID 2020 evaluation that features additional known activities for a known facility.
Openmfc: OPen media forensics challenge
The Open Media Forensics Challenge (OpenMFC) is a media forensics evaluation to facilitate development of systems that can automatically detect and locate manipulations in imagery (i.e., images and videos).
Building the Future: Generative models
The measurement and evaluation of AI systems involves collecting large amounts of labeled test data in order to assess system performance. The projects goal is to be able to use a limited amount of labeled data to more effectively measure system performance and robustness, where effectiveness is measured by the ability to produce confidence intervals that match the confidence intervals produced using large-scale evaluation datasets.
Evaluating scores of machine learning systems typically involves collecting a large amount of labeled data, which can be costly or impractical. The goal of active evaluation is to develop methods that minimize the amount of labeled data needed to determine a system’s scored according to a performance metric (accuracy, precision, etc…).
Building the Future: Active evaluation
The Data-Driven Discovery of Models (D3M) program aims to develop automated model discovery systems that enable users with subject matter expertise but no data science background to create empirical models of real, complex processes. This capability will enable subject matter experts to create empirical models without the need for data scientists, and will increase the productivity of expert data scientists via automation.
Data-driven discovery of models
Publications
G. Awad, A. Butt, K. Curtis, J. Fiscus, A. Godil, Y. Lee, A. Delgado, J. Zhang, E. Godard, B. Chocot -- Information Access Division, National Institute of Standards and Technology, USA
G. Awad, A. Butt, K. Curtis, Y. Lee, J. Fiscus, A. Godil, A. Delgado, J. Zhang, E. Godard, L. Diduch -- Information Access Division, National Institute of Standards and Technology, USA