DHS · DHS-2384 · record R1919

Verification Match Model

Department of Homeland Security · USCIS

Overview

Development stage: DeployedSource: c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.
High-impact designation: High-impactSource: a) High-impact
Impact justification: Not reported
Start date: 2024-05-22 (day precision)Source: 2024-05-22T00:00:00
Withheld from public reporting: Not reported

Mission

Topic area: Administrative Functions
Operational functions: Claims and case processing Derived by keyword rules; see methodology

Problem the AI is intended to solve

By consolidating these into a single, unified Verification Match Model within a separate microservice, the use case aims to improve the accuracy of responses and reduce the need for manual review. ML plays a key role in the continuous improvement of these models, ultimately reducing the need for manual case reviews.

Expected benefits

Leveraging AI in the USCIS verification matching process of known records across systems is beneficial because it streamlines existing USCIS review by 1) improving associated system accuracy, 2) reducing human-error by automating person-and-record match scoring, and 3) matching at a higher volume than traditional tools or manual processes can capably achieve.

System outputs

A recommendation and score that indicates person-and-record match probability used by verification systems (E-verify and SAVE) to improve accuracy in initial system response

Technology

AI classification: Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…
System name(s): Verification Information System
Custom-developed code: Yes
Public source code: Not reported

Sourcing

How it was built: Contract and in-houseSource: c) Developed with both contracting and in-house resources
Vendor (as reported): IBM
Vendors (standardized): IBM

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII: Yes
Privacy Impact Assessment: www.dhs.gov ↗
Authorization to Operate: Yes
Demographic features: Not reported
Training and evaluation data: Individual's Names, Dates of Birth, and Document Identifiers from USCIS sourced data contained in CIS2, C3, ELIS, and Global. These are all private datasets within USCIS.
Federal Data Catalog: Not reported

Governance

8 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Pre-deployment testingReported in placeSource: a) Yes

AI impact assessmentReported in progressSource: b) In-progress

Independent reviewReported in progressSource: d) In-progress

Ongoing monitoringReported in placeSource: a) Yes, sufficient monitoring protocols have been established

Operator trainingReported in progressSource: b) Establishment of sufficient and periodic training is in-progress

Fail-safeReported in placeSource: a) Yes

Appeal processReported in placeSource: a) Yes, an appropriate appeal process has been established

User and public consultationReported in progressSource: e) In-progress

Potential impacts and how they were identified

Model-match performance in terms of accuracy, precision, and recall. Identified via model evaluation and analysis for these performance statistics.