Scholarlink

Nutrition and farmer livelihoods are at risk due to a number of issues facing agriculture such as irregular crop yields, vulnerability to disease and the effects of weather variability. This research offers an AI-powered solution to these problems combining deep learning, advanced machine learning techniques and immediate data analysis to maximize agricultural yields, identify diseases early and offer useful insights on Crop Yield Prediction Dataset and PlantVillage Dataset. A complex approach is used in the system: forecasting the weather using the PyOWM API to forecast temperature, humidity and conditions for the next five days; crop disease detection using data augmentation and deep learning models like CNN (accuracy 99.14%), DenseNet201 (accuracy 99.04%) and VGG19 (accuracy 97%), and crop yield prediction using models like MLP (R2 Score: 0.8242), MLP + Regressor achieving robust accuracy through metrics like RMSE (0.2913) and MAE (0.1789) and RandomForest Regressor achieving maximum R2 Score (0.921). For real-time support the technology includes an AI chatbot that offers farmers recommendations, disease control techniques and customized suggestions. This project integrates models with high rates of accuracy to deliver an AI-driven system for weather forecasting, disease detection, yield prediction and real-time help via a chatbot. Telugu, Hindi and English versions of the user- friendly Streamlit UI are provided and SQLite manages the safe login and registration process.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Hybrid AI Chatbot for Crop Yield Optimization and Disease Prevention Using Deep Learning Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Jayaprada Somala,
  • Bhargavi Aggunna,
  • Hema Naga Sri Gajula,
  • Harika Jadam,
  • Vagdevi Marisetti

摘要

Nutrition and farmer livelihoods are at risk due to a number of issues facing agriculture such as irregular crop yields, vulnerability to disease and the effects of weather variability. This research offers an AI-powered solution to these problems combining deep learning, advanced machine learning techniques and immediate data analysis to maximize agricultural yields, identify diseases early and offer useful insights on Crop Yield Prediction Dataset and PlantVillage Dataset. A complex approach is used in the system: forecasting the weather using the PyOWM API to forecast temperature, humidity and conditions for the next five days; crop disease detection using data augmentation and deep learning models like CNN (accuracy 99.14%), DenseNet201 (accuracy 99.04%) and VGG19 (accuracy 97%), and crop yield prediction using models like MLP (R2 Score: 0.8242), MLP + Regressor achieving robust accuracy through metrics like RMSE (0.2913) and MAE (0.1789) and RandomForest Regressor achieving maximum R2 Score (0.921). For real-time support the technology includes an AI chatbot that offers farmers recommendations, disease control techniques and customized suggestions. This project integrates models with high rates of accuracy to deliver an AI-driven system for weather forecasting, disease detection, yield prediction and real-time help via a chatbot. Telugu, Hindi and English versions of the user- friendly Streamlit UI are provided and SQLite manages the safe login and registration process.