Dato 2025
Publikasjonstype Vitenskapelig tidsskriftpublikasjon
An Introduction to prismAId: Open-Source and Open Science AI for Advancing Information Extraction in Systematic Reviews
Publikasjonsdetaljer
Publikasjonstype: Vitenskapelig artikkel
Tidsskrift: Journal of Open Source Software (JOSS)
Volum: 10
Artikkelnummer: 7616
Utgave: 108
Original versjon/DOI: https://doi.org/10.21105/joss.07616
Sammendrag:
prismAId is an open-source tool designed to streamline systematic literature reviews by leveraging generative AI models for information extraction. It offers an accessible, efficient, and replicable method for extracting and analyzing data from scientific literature, eliminating the need for coding expertise. Supporting various review protocols, including PRISMA 2020, prismAId is distributed across multiple platforms – Go, Python, Julia, R – and provides user-friendly binaries compatible with Windows, macOS, and Linux. The tool integrates with leading large language models (LLMs) such as OpenAI’s GPT series, Google’s Gemini, Cohere’s Command, and Anthropic’s Claude, ensuring comprehensive and up-to-date literature analysis. prismAId facilitates systematic reviews, enabling researchers to conduct thorough, fast, and reproducible analyses, thereby advancing open science initiatives.