Explorations in Data Innovations: Can Machine Learning Support Data Catalog Development? - NF
About the Study
In 2021, the Chief Evaluation Office (CEO) funded contractor Westat Insight and their partner American Institutes for Research to conduct the Explorations in Data Innovations project under the Administrative Data Research and Analysis portfolio of studies. This use case study was designed to explore options around employing machine learning to create and maintain a public-facing, labor-related data catalog. Data catalogs are one way to address a goal of CEO to make data sources and initiatives in the employment and training field more accessible to researchers. A publicly available data catalog would help labor researchers more easily understand the full range of available labor-related datasets that could potentially answer pressing questions on labor-related policies and programs. The development and maintenance of data catalogs is labor-intensive; this project’s goal was to determine the feasibility of different automation options for building data catalogs. In this effort, the study team explored relevant literature, piloted a manual data catalog assembly process, developed options for each step of the data catalog process, and consulted with a technical working group of computer science experts.
This Department of Labor-funded study contributes to the labor evidence-base to inform data, methods, and tools that build evidence on Departmental programs and policies and addresses Departmental strategic goals and priorities.
- To what extent can automated machine-learning algorithms be used to identify new data sources and initiatives on employment and training outcomes?
- Explorations in Data Innovations: Can Machine Learning Support Data Catalog Development? (Final Report, July 2024)
The Department of Labor’s (DOL) Chief Evaluation Office (CEO) sponsors independent evaluations and research, primarily conducted by external, third-party contractors in accordance with the Department of Labor Evaluation Policy and CEO’s research development process.