Development and Evaluation of a Generative AI Chatbot for Database Searching in Systematic Review

Key Points

• The study aims to evaluate the effectiveness of a generative AI chatbot in facilitating database searches for systematic reviews.
• Developed a generative AI chatbot for systematic review searches.
• Assessed the chatbot's search identification rate (67%-72%).
• Integrated librarian expertise and structured prompts to refine search accuracy.
• Chatbot achieved an identification rate of 67%-72% in systematic review searches.
• Expert involvement and controlled vocabularies may enhance chatbot performance.
• Further benchmarking is necessary for proper guidance on adoption.