Internal Referral Module (IRM)

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

Presumed high-impact, determined notSource: b) Presumed high-impact, but determined not high impact

Impact justification

The output from this AI use case does not drive any agency decision, including the categories listed in section 6 of the memorandum. The output is a recommendation to assign a grant application to the appropriate Agency staff based on the scientific content of the application. The person to whom the grant application has been referred to can accept, reject, or reassign the application based on their expertise.

Start date

2023-02 (month precision)Source: Feb-23

Withheld from public reporting

NoSource: a) No

Mission

Topic area

Administrative Functions

Operational functions

No function tag matched Derived by keyword rules; see methodology

Problem the AI is intended to solve

Automated Assignment of Grants to Program Officers

Expected benefits

The original IRM application grew out of a desire to refer applications to the appropriate Program Officer to manage the scientific research that fit their portfolio.  This manual referral of grant applications still exists within IRM and has been complemented by use of AI/NLP capabilities.

System outputs

The outputs are referrals to Program Officers, Program Class Codes, Organizational units - Divisions and Branches and Scientific Research Clusters.

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)

eRA

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)

Leidos and Highrise

Vendors (standardized)

Leidos

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

Data and privacy

Involves PII

No

Privacy Impact Assessment

Not reported

Authorization to Operate

Yes

Demographic features

k) None of the above

Training and evaluation data

We use eRA grant application data for all fine tuning and optimization for the models. Specially, we extract the title, abstract, specific aims and public health narrative to train our models for prediction.

Federal Data Catalog

Not reported

Governance

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

Potential impacts and how they were identified

Not reported