Loading…

KNIMEZoBot: Enhancing Literature Review with Zotero and KNIME OpenAI Integration using Retrieval-Augmented Generation

Academic researchers face challenges keeping up with exponentially growing published findings in their field. Performing comprehensive literature reviews to synthesize knowledge is time-consuming and labor-intensive using manual approaches. Recent advances in artificial intelligence provide promisin...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2023-11
Main Authors: Alshammari, Suad, Basalelah, Lama, Walaa Abu Rukbah, Alsuhibani, Ali, Wijesinghe, Dayanjan S
Format: Article
Language:English
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Academic researchers face challenges keeping up with exponentially growing published findings in their field. Performing comprehensive literature reviews to synthesize knowledge is time-consuming and labor-intensive using manual approaches. Recent advances in artificial intelligence provide promising solutions, yet many require coding expertise, limiting accessibility. KNIMEZoBot represents an innovative integration of Zotero, OpenAI, and the KNIME visual programming platform to automate literature review tasks for users with no coding experience. By leveraging KNIME's intuitive graphical interface, researchers can create workflows to search their Zotero libraries and utilize OpenAI models to extract key information without coding. Users simply provide API keys and configure settings through a user-friendly interface in a locally stored copy of the workflow. KNIMEZoBot then allows asking natural language questions via a chatbot and retrieves relevant passages from papers to generate synthesized answers. This system has significant potential to expedite literature reviews for researchers unfamiliar with coding by automating retrieval and analysis of publications in personal Zotero libraries. KNIMEZoBot demonstrates how thoughtfully designed AI tools can expand accessibility and accelerate knowledge building across diverse research domains.
ISSN:2331-8422