Scholarlink

<p>Accurate rainfall prediction is critical for ensuring agricultural resilience, disaster preparedness, and energy resource planning, particularly in monsoon-driven regions like West Bengal, India. This study introduces an efficient machine learning framework that leverages ensemble techniques for high-accuracy annual precipitation forecasting. The methodology utilizes historical climate data from NASA’s Power Data Access Viewer (1984–2022), incorporating features such as temperature, humidity, wind speed, and previous rainfall. An extensive data preprocessing pipeline was employed, including label encoding, outlier detection, normalization, dimensionality reduction using Principal Component Analysis (PCA), and class balancing via SMOTE and SMOTE-Tomek. Feature selection was performed using correlation analysis, chi-square testing, wrapper methods, and embedded models. Multiple ensemble approaches, bagging, boosting, and stacking, were benchmarked using tenfold cross-validation. The highest accuracy of 99.38% was achieved using a two-layer stacking ensemble combining Extreme Gradient Boosting (XGBoost) and Extra Trees Classifier (ETC). This work demonstrates the practical effectiveness of integrated machine learning models for environmental prediction and showcases a scalable framework for real-world climate-driven data forecasting.</p>

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

Efficient Ensemble Approach Towards the Betterment of Annual Precipitation Prediction in West Bengal, India

  • 期刊论文

  • 发表日期：

    2026年7月17日

查看全文
  • Priyabrata Sain,
  • Sayan Pal,
  • Rahul Karmakar

摘要

Accurate rainfall prediction is critical for ensuring agricultural resilience, disaster preparedness, and energy resource planning, particularly in monsoon-driven regions like West Bengal, India. This study introduces an efficient machine learning framework that leverages ensemble techniques for high-accuracy annual precipitation forecasting. The methodology utilizes historical climate data from NASA’s Power Data Access Viewer (1984–2022), incorporating features such as temperature, humidity, wind speed, and previous rainfall. An extensive data preprocessing pipeline was employed, including label encoding, outlier detection, normalization, dimensionality reduction using Principal Component Analysis (PCA), and class balancing via SMOTE and SMOTE-Tomek. Feature selection was performed using correlation analysis, chi-square testing, wrapper methods, and embedded models. Multiple ensemble approaches, bagging, boosting, and stacking, were benchmarked using tenfold cross-validation. The highest accuracy of 99.38% was achieved using a two-layer stacking ensemble combining Extreme Gradient Boosting (XGBoost) and Extra Trees Classifier (ETC). This work demonstrates the practical effectiveness of integrated machine learning models for environmental prediction and showcases a scalable framework for real-world climate-driven data forecasting.