Abstract:
|
The Office of Immigration Statistics (OIS) Enforcement Lifecycle Tool matches individual?level records across Department of Homeland Security (DHS) and Department of Justice (DOJ) data systems to analyze how aliens move through the immigration enforcement process. OIS implemented a rule-based methodology that sorts records based on a set number of key identifiers and compares records in a specific distance from a given record. At each data ingest, OIS stacks source datasets into a single file (the Flow data), and then uses a matching algorithm that iteratively sorts records on pairs of person? and event?level identifiers. Following each sort, OIS’ matching algorithm analyzes sequential records and assigns a shared ID to matching records. Matched records are kept together, and the data are re?sorted on a second pair of identifiers before repeating the process for each set of in?scope identifiers. Currently, OIS re?matches all source datasets each time new records are ingested, yielding a new set of unique person identifiers. OIS is updating its methodology to speed processing time and maintain consistent person identifiers matching new records against already?merged data.
|