Crop Yield Prediction and Disease Detection

Abstract

Agriculture plays a vital role in ensuring food security and economic sustainability, yet it faces significant challenges such as climate variability, pest infestations, and plant diseases that adversely affect crop productivity. This paper proposes an intelligent agricultural management system that integrates machine learning and deep learning techniques for crop yield prediction, plant disease detection, and crop recommendation. A hybrid approach using SARIMAX and XGBoost models is employed for crop yield prediction by capturing both temporal patterns and nonlinear relationships in environmental and historical data, including rainfall, temperature, and location-specific factors. For disease detection, YOLOv5 is utilized to accurately identify and localize diseases in leaf images, while a Large Language Model (Groq) is used to generate detailed explanations and treatment recommendations.The system incorporates a weather analysis and crop recommendation module that provides short-term weather forecasts and suggests suitable crops based on environmental conditions. A user-friendly chatbot powered by a Large Language Model enables farmers to interact with the system, upload images, and receive real-time guidance. The proposed system offers a scalable and automated solution for precision agriculture, enhancing decision-making, reducing crop losses, and promoting sustainable farming practices.

Introduction

This text describes the development of an AI-based smart agriculture system designed to improve crop productivity, disease management, and farming decision-making using machine learning, deep learning, and large language models.

Agriculture faces major challenges such as climate change, soil degradation, pest attacks, and plant diseases, while traditional farming methods are slow, manual, and error-prone. To address these issues, the proposed system integrates multiple intelligent modules to support farmers with real-time insights and automation.

The system combines:

• Crop yield prediction using a hybrid model (SARIMAX for time-series patterns and XGBoost for nonlinear relationships)
• Plant disease detection using YOLOv5 on leaf images, followed by an LLM (via Groq) for explanations and treatment suggestions
• Weather analysis and crop recommendation based on short-term forecasts and environmental conditions
• A chatbot interface powered by a large language model for farmer interaction and guidance

The literature survey shows that machine learning and deep learning models like Random Forest, XGBoost, CNNs, and ensemble methods have achieved high accuracy in agriculture tasks such as yield prediction and disease detection. However, most existing systems are limited because they focus on only one task (either prediction or disease detection), lack real-time integration, and are not fully accessible to farmers.

The proposed system addresses these gaps by building a unified platform that combines all key agricultural functions into one system. It uses a web-based interface where farmers can upload images, input crop details, receive predictions, view weather updates, and interact with a chatbot.

Conclusion

This paper presents an intelligent agricultural management system that integrates machine learning, deep learning, and natural language processing techniques to address key challenges in modern farming. The proposed system combines crop yield prediction, plant disease detection, weather analysis, and crop recommendation within a unified framework, providing a comprehensive solution for precision agriculture. The yield prediction module, based on SARIMAX and XGBoost, effectively captures both temporal trends and nonlinear relationships in environmental and historical data, resulting in accurate and reliable forecasts. “The yield prediction model achieved an R2 score of 0.93, indicating 93% variance explanation in crop yield prediction.” The disease detection module, utilizing YOLOv5, demonstrates high accuracy in identifying and localizing plant diseases, enabling early diagnosis and timely intervention. In addition, the integration of a Large Language Model (LLM) via the Groq platform enhances the system by generating detailed explanations, treatment recommendations, and enabling chatbot-based interaction. The system achieved a disease detection accuracy of 97.66%, indicating strong performance and practical applicability. Overall, the proposed framework highlights the effectiveness of combining computer vision, time-series analysis, and machine learning techniques for intelligent agriculture. By reducing dependency on manual monitoring and supporting data-driven decision-making, the system contributes to improved crop productivity, efficient resource utilization, and sustainable farming practices.

References

[1] R. Screpnik, E. Zamudio and L. I. Gimenez, \"Artificial Intelligence in Agriculture: A Systematic Review of Crop Yield Prediction and Optimization,\" in IEEE Access, vol. 13, pp. 70691-70697, 2025, doi: 10.1109/ACCESS.2025.3560631. [2] P. Sharma, P. Dadheech, N. Aneja and S. Aneja, \"Predicting Agriculture Yields Based on Machine Learning Using Regression and Deep Learning,\" in IEEE Access, vol. 11, pp. 111255-111264, 2023, doi: 10.1109/ACCESS.2023.3321861. [3] R. Rani, J. Sahoo, S. Bellamkonda, S. Kumar and S. K. Pippal, \"Role of Artificial Intelligence in Agriculture: An Analysis and Advancements With Focus on Plant Diseases,\" in IEEE Access, vol. 11, pp. 137999-138019, 2023, doi: 10.1109/ACCESS.2023.3339375. [4] A. Badshah, B. Yousef Alkazemi, F. Din, K. Z. Zamli and M. Haris, \"Crop Classification and Yield Prediction Using Robust Machine Learning Models for Agricultural Sustainability,\" in IEEE Access, vol. 12, pp. 162799-162813, 2024, doi: 10.1109/ACCESS.2024.3486653. [5] M. Ashfaq, I. Khan, A. Alzahrani, M. U. Tariq, H. Khan and A. Ghani, \"Accurate Wheat Yield Prediction Using Machine Learning and Climate-NDVI Data Fusion,\" in IEEE Access, vol. 12, pp. 40947-40961, 2024, doi: 10.1109/ACCESS.2024.3376735. [6] A. A. Mamun, D. Ahmedt-Aristizabal, M. Zhang, M. Ismail Hossen, Z. Hayder and M. Awrangjeb, \"Plant Disease Detection Using Self-Supervised Learning: A Systematic Review,\" in IEEE Access, vol. 12, pp. 171926-171943, 2024, doi: 10.1109/ACCESS.2024.3475819. [7] S. Kumar, M. Pant and A. Nagar, \"Forecasting the Sugarcane Yields Based on Meteorological Data Through Ensemble Learning,\" in IEEE Access, vol. 12, pp. 176539-176553, 2024, doi: 10.1109/ACCESS.2024.3502547. [8] U. Shafi et al., \"Tackling Food Insecurity Using Remote Sensing and Machine Learning-Based Crop Yield Prediction,\" in IEEE Access, vol. 11, pp. 108640-108657, 2023, doi: 10.1109/ACCESS.2023.3321020. [9] S. T. Haider et al., \"An Ensemble Machine Learning Framework for Cotton Crop Yield Prediction Using Weather Parameters: A Case Study of Pakistan,\" in IEEE Access, vol. 12, pp. 124045-124061, 2024, doi: 10.1109/ACCESS.2024.3454511. [10] M. Khalid and M. A. Talukder, \"A Hybrid Deep Multistacking Integrated Model for Plant Disease Detection,\" in IEEE Access, vol. 13, pp. 116037-116053, 2025, doi: 10.1109/ACCESS.2025.3583793. [11] M. J. Hoque et al., \"Incorporating Meteorological Data and Pesticide Information to Forecast Crop Yields Using Machine Learning,\" in IEEE Access, vol. 12, pp. 47768-47786, 2024, doi: 10.1109/ACCESS.2024.3383309. [12] R. Mai, Q. Xin, J. Qiu, Q. Wang and P. Zhu, \"High Spatial Resolution Soil Moisture Improves Crop Yield Estimation,\" in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 17, pp. 19067-19077, 2024, doi: 10.1109/JSTARS.2024.3417424. [13] R. M. Saleem et al., \"Internet of Things Based Weekly Crop Pest Prediction by Using Deep Neural Network,\" in IEEE Access, vol. 11, pp. 85900-85913, 2023, doi: 10.1109/ACCESS.2023.3301504. [14] H. Najjar, M. Miranda, M. Nuske, R. Roscher and A. Dengel, \"Explainability of Subfield Level Crop Yield Prediction Using Remote Sensing,\" in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 18, pp. 4141-4161, 2025, doi: 10.1109/JSTARS.2025.3528068. [15] S. Iniyan and R. Jebakumar, \"Phenotype Based Smart Mobile Application for Crop Yield Prediction and Forecasting Using Machine Learning and Time Series Models,\" in Journal of Mobile Multimedia, vol. 18, no. 3, pp. 603-634, May 2022, doi: 10.13052/jmm1550-4646.1837. [16] A. Reyana, S. Kautish, P. M. S. Karthik, I. A. Al-Baltah, M. B. Jasser and A. W. Mohamed, \"Accelerating Crop Yield: Multisensor Data Fusion and Machine Learning for Agriculture Text Classification,\" in IEEE Access, vol. 11, pp. 20795-20805, 2023, doi: 10.1109/ACCESS.2023.3249205. [17] B. Ramana Reddy, G. Kalnoor, M. Devashish and P. Sai Karthik Reddy, \"Deep Learning Based Mobile Application for Automated Plant Disease Detection,\" in IEEE Access, vol. 13, pp. 107917-107925, 2025, doi: 10.1109/ACCESS.2025.3581099.

Copyright

Copyright © 2026 Akshara Vinod, Fahad N N, Megha K P, Saniga Sudheer, Binu Balakrishnan. 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.