Personalized AI Chatbot for Indian Agriculture: Soil, Crop, and Weather-Based Recommendations

Abstract

Agriculture in India is highly dependent on various parameters such as soil management, crop selection irrigation techniques and weather. However the marginal farmers in India face numerous challenges which includes low knowledge of the cropfertilizers, crop diseases, limited expert advices, language barriers and low digital literacy. The research proposes a Personalized AI Chatbot for Indian Agriculture that delivers real-time, multilingual, and context aware recommendations based on soil parameters, crop stage, crop diseases and weather forecasts. The system integrates with models like Natural Language Processing (NLP), machine learning classifiers, and open environmental APIs help generate personalized responses and advisories in English along with the regional languages such as Hindi and Marathi. An experimental evaluation revealed that the chatbot achieved an overall functional accuracy of 86.6%, with the Random Forest model achieving 92% predictive accuracy for soil- and crop-based recommendations. Multilingual processing demonstrated strong translation quality (BLEU scores between 0.71 and 0.78), while response times remained low (1.42–2.95 seconds), supporting real-time interaction even under rural network constraints. User feedback confirmed high satisfaction and usability. The results show that the proposed system can effectively bridge the digital and informational gaps in Indian agriculture, offering a scalable, accessible, and farmer-centric AI solution.

Introduction

The text discusses the need for an AI-powered multilingual agricultural chatbot for Indian farmers that provides personalized, real-time farming advice.

Key Problem

Agriculture in India is dominated by small and marginal farmers who face:

• Fragmented land holdings
• Limited access to expert agronomic advice
• Low digital literacy and language barriers
• Uneven access to reliable, localized agricultural information

Although internet usage is growing rapidly in rural India, most existing agritech tools are not designed for regional languages, low literacy users, or farm-specific recommendations.

Limitations of Existing Systems

Current agricultural advisory systems and chatbots:

• Provide generalized rather than farm-specific advice
• Lack integration of soil data, weather forecasts, and crop health analysis
• Have weak support for regional languages and dialects
• Often rely on static datasets instead of real-time updates
• Fail to work well in low-connectivity rural environments

Even advanced AI chatbots and government-supported platforms (like Farmer.Chat, AgriFriend, and CropCare Companion) struggle with personalization, offline usability, and full environmental integration.

Government Efforts

India has made progress through initiatives like:

• Digital India and BharatNet (rural connectivity)
• PM-Kisan (farmer database support)
• eNAM (digital agricultural markets)
• AgriStack (integrated agricultural data ecosystem)

However, these systems still lack strong farmer-facing intelligent advisory tools.

Proposed Solution

The paper proposes a personalized AI chatbot for agriculture that:

• Provides crop advice, soil health insights, and disease detection
• Integrates weather, soil, and farm-specific data
• Uses multilingual text and voice-based interaction
• Employs RAG (Retrieval-Augmented Generation) and NLP for accurate responses

This system aims to solve three main issues:

1. Lack of personalized, plot-specific guidance
2. Language and literacy barriers
3. Poor connectivity and limited digital accessibility

Literature Review Summary

Research shows steady progress in:

• AI-based agricultural decision systems
• Multilingual chatbots
• IoT-enabled farming tools
• Geospatial and weather-integrated advisory systems

However, most systems still fail to achieve full integration of real-time data, personalization, multilingual support, and rural usability together.

Conclusion

This research resulted in the development of a personalized, multilingual AI-powered agricultural guidance chatbot designed to assist Indian farmers with real-time, context-aware, and effortlessly accessible recommendations. By combining soil characteristics, crop requirements, and weather forecasts with transformer-based NLP models, the system effectively addressed critical limitations in existing solutions, such as minimal personalization, inadequate multilingual support, and the lack of real-time environmental data integration. Experimental evaluations yielded strong performance metrics, with the chatbot achieving 86.6% functional accuracy, 92% model accuracy via Random Forest classification, high-quality multilingual translation (BLEU scores between 0.71 and 0.78), and low average response times, which are suitable for rural connectivity. User evaluations confirmed the system\'s usability, with high satisfaction levels and positive remarks on the voice-enabled, language-inclusive interface. The aggregate findings reinforce the system\'s technical efficiency as well as its social relevance in the Indian agricultural environment. The methodology has the potential to improve smallholder farmers\' access to expert guidance by simplifying decision-making and increasing information accessibility. Despite of its attributes, there are various chances for improvement. Future initiatives could include real-time soil sensor networks (IoT) to improve suggestion accuracy and reduce dependability on manually entered soil data. The use of powerful generative AI models geared specifically to Indian agricultural datasets could improve interaction quality, allowing for more deeper and adaptive interactions. Expanding assistance to encompass more regional languages and dialects would improve accessibility for farmers from various linguistic backgrounds. Furthermore, the application of predictive analytics for pest outbreaks, yield projections, and climate risk warnings can elevate the system from an advisory function to comprehensive decision support. Finally, large-scale field trials involving farmers from several states will help evaluate results under real-world situations and inform incremental improvements. In conclusion, the suggested chatbot makes a significant contribution in bridging the digital divide between Indian Rural Agricultural and urban cities by combining artificial intelligence, multilingual NLP, and environmental data into a single, farmer-centric platform. This technology has a lot of potential to become a scalable digital agriculture tool that can support millions of farmers nationwide.

References

[1] Government of India, Ministry of Agriculture & Farmers Welfare, Agricultural Statistics at a Glance 2023, Govt. of India, 2023. [2] Telecom Regulatory Authority of India (TRAI), Telecom Performance Indicators Report, Mar. 2025. [3] Internet & Mobile Association of India (IAMAI) and Kantar, Internet in India Report 2024, IAMAI/Kantar, Jan. 2025. [4] Ministry of Electronics and Information Technology, Pradhan Mantri Gramin Digital Saksharta Abhiyan (PMGDISHA) Dashboard, Govt. of India, 2025. [5] Ministry of Agriculture & Farmers Welfare, eNAM Progress Report, Govt. of India, 2024. [6] Ministry of Agriculture & Farmers Welfare, AgriStack: Building India’s Digital Agriculture Backbone, Govt. of India, 2024. [7] Ministry of Agriculture & Farmers Welfare, PM-Kisan Dashboard, Govt. of India, 2024. [8] Digital Green, Farmer.Chat Deployment in Andhra Pradesh – Technical Brief, Digital Green Foundation, 2024. [9] E. Asolo and R. Nair, “AI-Powered Decision Support Systems for Sustainable Agriculture Using AI Chatbot Solution,” International Journal of Agricultural Informatics, vol. 12, no. 3, pp. 83–90, 2024. [10] A. Sable, M. Patil, and V. Joshi, “CropCare Companion: Multilingual Conversational AI for Indian Farmers,” Procedia Computer Science, vol. 224, pp. 33–42, 2025. [11] R. Kumar and P. Singh, “E-AGRO: Intelligent Chatbot Using IoT and AI to Enhance Farming Industry,” International Journal of Engineering Science, vol. 11, no. 8s, pp. 59–66, 2023. [12] P. Kulkarni and A. Mane, “AgriFriend: Conversational AI Chatbot Using Open Source Frameworks,” International Journal of Research in Engineering, Science and Management (IJRESM), vol. 4, no. 4, pp. 76–82, 2022. [13] S. Sharma et al., “A Machine Learning-Based Mobile Chatbot for Crop Farmers,” Proceedings of International Conference on Information and Communication Technology (ICICT), pp. 245–251, 2022. [14] V. Patange et al., “AGEOChatbot: Geospatial Decision Support for Smart Farming,” Agricultural Informatics Journal, vol. 18, no. 2, pp. 90–98, 2023. [15] N. Rao and D. Patel, “Scaling AI-Powered Agricultural Services,” IEEE Access, vol. 13, pp. 11785–11793, 2025. [16] Farmer.Chat Consortium, “Context-Aware Generative AI Chatbot for Farmers,” Technical Report, 2024. [17] AgroBuddy Team, Empowering Indian Farmers Through Precision Farming Chatbot, Research Report, 2023. [18] S. R. Verma et al., “AI-Based Multilingual Chatbot for Agriculture Assistance,” Proceedings of IEEE Conference on Innovation in Computing, pp. 110–118, 2024. [19] ICAR (Indian Council of Agricultural Research), KrishiBot: AI Chatbot Pilot for KVK Farmers, Govt. of India, 2023. [20] IFFCO Kisan, AI-Powered Kisan Chatbot, IFFCO Kisan Portal, 2023. [21] NITI Aayog, KisanGPT Pilot on Generative AI for Agriculture, Govt. of India, 2024. [22] CRISIL Ratings, Internet Penetration in Rural India Likely to Grow by 4–5% by FY26, Analytical Report, 2025.

Copyright

Copyright © 2026 Vedant Patil, Vaidehi Gavali, Rohit Katkade Katkade , Prachi Bhure. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.