Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 452 - Data Matching Practice and Application in the United States Immigration System
Type: Topic Contributed
Date/Time: Thursday, August 6, 2020 : 10:00 AM to 11:50 AM
Sponsor: Government Statistics Section
Abstract #313361
Title: Data Matching in Immigration Enforcement Lifecycle Tool
Author(s): Hongwei Zhang*
Companies: Office of Immigration Statistics, DHS
Keywords: data matching; entity resolution; immigration data; enforcement; life cycle; statistical reporting
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.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program