AI Research Assistant in Dryland Agriculture - Retrieval Augmented Generation Based Chat Bot for Literature Review
2026
Patil, Mukund | Rupavatharam, Srikanth | Mitnala, Sreevani | Katakamsetti, Nagendra | Reddy, Nagarjuna N.
Dryland agriculture faces persistent challenges in accessing timely, reliable, and context-specific scientific knowledge due to fragmented literature, manual search limitations, and inadequate understanding of relationships between research findings and real-world applications. This brochure presents the AI Research Assistant in Dryland Agriculture, a Retrieval Augmented Generation (RAG)–based chatbot designed to transform static agricultural literature into an interactive, evidence-grounded digital assistant. Built on ICRISAT’s Open Access Repository of over 12,000 validated publications, the system integrates automated document processing, semantic embedding, vector-based retrieval, and open-source large language models to deliver accurate, context-aware responses to natural-language queries. Multi-layered quality assurance and transparent source attribution ensure scientific reliability, while an accessible chat and voice-enabled interface broadens usability for researchers, extension workers, and practitioners. Early results demonstrate superior performance over traditional keyword searches and improved synthesis of complex knowledge across documents. The platform represents a significant advancement in agricultural knowledge management, supporting evidence-based decision-making and accelerating innovation in dryland agriculture.
اظهر المزيد [+] اقل [-]الكلمات المفتاحية الخاصة بالمكنز الزراعي (أجروفوك)
المعلومات البيبليوغرافية
تم تزويد هذا السجل من قبل International Livestock Research Institute