Gå till huvudinnehållet

Using AI to identify and review research

AI-aided tools for systematic reviews

A systematic review is a method in research that involves identifying and synthesizing all available existing research on a specific topic, to provide a comprehensive overview and ensure transparency and replicability. Currently there are no agreed upon methodology for using AI-aided tools in the systematic review process other than screening articles for inclusion in the systematic review.

Screening prioritization

Systematic reviews require extensive literature searches followed by title/abstract screening, data extraction, and quality assessment. During the title/abstract screening phase reviewers need to screen a large amount of studies for inclusion. A well-established approach to increase the screening efficiency is screening prioritization. Screening prioritization uses machine learning to change the order of the records to be screened, from a random to a more intelligent order, as the reviwer decides which records to include or exclude.

AI-aided tools for screening and study selection

Acknowledging use of AI-aided tools

For transparency and reproducability acknowledge your use of AI-aided tools. Here are some suggestions on how to acknowledge the use of AI in academic research:

  • Cite it in the text and reference list of your work
  • Describe your use as part of your methodology
  • Include an appendix with screenshots or transcripts of prompts and AI-generated responses

 

Citing use of AI-aided tools

Publishers may have different policies on whether or not AI is allowed, and how to cite it. Check your publisher's information for authors webpage, or contact their editorial staff, for details.

APA 7 reference for ChatGPT

OpenAI. (Year). ChatGPT (Month Day version) [Large language model]. https://chat.openai.com