Data Mining in Federal Agencies Rockville, MD, and University of Pittsburgh, Pittsburgh, PA, USA

CONTENTS 32.1 Data Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529 32.2 Indexing Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534 32.3 Screening for Structure with Locally Low Dimension . . . . . . . . . . . . . . . . . . . . . 537 32.4 Estimating Exposure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 546

Federal agencies in the United States collect enormous quantities of data. Recently, they have begun to attempt to apply data mining to extract additional insights from their collections.