AI evidence extraction

A guide for structured literature reviews in business research: The state-of-the-art and how to integrate generative artificial intelligence

Fabian Tingelhoff, Micha Brugger, Jan Marco Leimeister · Journal of Information Technology · 2024

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

10/10

Relevance

1/4

Quality (LMQS)

Evidence

Citations

16.48

FWCI

Top 10%

Impact

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.1177/02683962241304105

Methodology & findings

Study design

Narrative literature review with criteria-centric analysis of SLR methodologies and quality standards; synthesis of extant processes to develop a unified framework for integrating generative AI.

Main result

The study synthesizes established state-of-the-art processes and quality standards in structured literature reviews (SLRs) and "delineate[s] the specific scenarios conducive to incorporating Gen.AI into this fundamental framework, as well as situations where its integration may not be suitable." The authors provide a criteria-centric approach that focuses on "what we should allow Gen.AI to do, irrespective of its capabilities."

Reports effect sizes.

Research paradigm

Interpretive/argumentative

Author conclusions

The authors conclude that their study "informs researchers in the art and science of SLRs" by providing both "a unified process and criterion set" that serves as "the foundational framework for integrating Gen.AI" and "a detailed, step-by-step guide—akin to a 'cooking recipe'—to effectively integrate Gen.AI in SLRs, ensuring adherence to established quality criteria."

Open questions raised

The authors identify that previous research focused on Gen.AI capabilities and limitations, but that "the rapid evolution of Gen.AI often outpaces the publication of methodological papers." They address the gap of determining what should be allowed rather than what can be done.
The paper identifies that previous research has predominantly focused on capabilities and limitations of Gen.AI rather than establishing ethical and methodological guidelines. The authors note that the rapid evolution of Gen.AI outpaces publication of methodological guidance.
The authors identify that previous research predominantly focused on Gen.AI capabilities and limitations, but the rapid evolution of Gen.AI often outpaces methodological guidance. They position their work to fill this gap by establishing criteria-based standards for integrating Gen.AI into SLRs.

Data: not_statedCode: not_statedExtracted from: pdfAgreement 75%

Explore related topics

Systematic Review Automation Literature Discovery Literature Review