Utilizing Public Data for Data Enhancement and Analysis of Federal Acquisition Data

Abstract

It is challenging to standardize data; yet, the capabilities to draw upon data across information systems hold huge potential for improving defense acquisition and procurement. Acquisition planning and management involves many decision-making and action-taking processes that cover a complex environment including actual acquisition, contracting, fiscal, legal, personnel, and regulatory requirements. A sound decision-making process has to rely on data high quality data. Often the available data is dirty, outdated, incomplete, or insufficient for the expert to make a decision. On the other hand, there are enormous amounts of data on the web that can be utilized to crystalize the needed information. These data repositories are often publicly accessible and from a variety of sources including websites, government reports, news, wikis, blogs, online forums, and social media. This paper investigates how to leverage the information in public data sources to complement the internal data in order to support effective acquisition planning and management. This research is based on publicly available government acquisition databases at usaspending.gov and fpds.gov. It takes a data science approach for analyzing acquisition databases and focuses on two major tasks: (1) research on leveraging the web data for quality assessment and improvement of federal acquisition data and (2) research on appropriate data analytic techniques to discover useful information that can potentially help federal acquisition management and planning process.