DOI · DOI-0106 · record R2650

Accelerating scientific discovery through AI-driven literature synthesis and meta-analysis using large language models [2024 INV#WO0000000201793]]

Department of the Interior · USGS

Overview

Development stage: In development or acquisitionSource: a) Pre-deployment – The use case is in a development or acquisition status.
High-impact designation: Not high-impactSource: c) Not high-impact
Impact justification: Not reported
Start date: Not reported
Withheld from public reporting: NoSource: a) No

Mission

Topic area: Science
Operational functions: No function tag matched Derived by keyword rules; see methodology

Problem the AI is intended to solve

A team of USGS researchers are conducting a review of literature on drought and its affects on the western United States. Due to a large volume of literature, we are using AI to facilitate an efficient and reliable literature synthesis that can inform decision-making. These include effectively ranking most relevant literature to questions, identifying gaps in the literature, identifying data gaps, and coalescing currently known information.

Expected benefits

These efforts will help expedite reviewing thousands of scientific studies that can support the project, but also support future efforts of compiling relevant information for NEPA planning or related tasks.

System outputs

The outputs from AI will include identifications of study areas, relatedness to multiple topics (drought characteristics, hydrological and ecological processes, and hydrological and ecological responses). Using multiple LLMs and deep learning methods to evaluate the same context, we will have comparative information for model ensembles.

Technology

AI classification: Natural language processingSource: Natural Language Processing: AI that processes, interprets, and shares information in human language.
System name(s): Not reported
Custom-developed code: No
Public source code: Code will be published at the end of the project and currently resides as private on USGS GitLab.

Sourcing

How it was built: Not reported
Vendor (as reported): Not reported
Vendors (standardized): None on the standard list

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: No
Demographic features: l) Other
Training and evaluation data: Not reported
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.

Pre-deployment testingNo answer reported

AI impact assessmentNo answer reported

Independent reviewNo answer reported

Ongoing monitoringNo answer reported

Operator trainingNo answer reported

Fail-safeNo answer reported

Appeal processNo answer reported

User and public consultationNo answer reported

Potential impacts and how they were identified

Not reported