Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
Human-led and artificial intelligence-automated critical appraisal of systematic reviews: Comparative evaluation
0
Zitationen
4
Autoren
2025
Jahr
Abstract
Large language models can support aspects of critical appraisal evidence but lack contextual reasoning and methodological insight required for complex judgments. While Claude 3.7 Sonnet aligned most closely with human reviewers, human oversight remains essential. Large language models should serve as adjuncts and not substitutes for evidence-based practice.
Ähnliche Arbeiten
The PRISMA 2020 statement: an updated guideline for reporting systematic reviews
2021 · 84.856 Zit.
Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement
2009 · 82.787 Zit.
The Measurement of Observer Agreement for Categorical Data
1977 · 76.851 Zit.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement
2009 · 62.738 Zit.
Measuring inconsistency in meta-analyses
2003 · 61.458 Zit.