Scholarlink

A direct relationship exists between methane (CH \(_{4}\) ) emissions from rice paddies and their environmental impacts that is already recognised in the scientific community. Changes in rainfall patterns and the increasing frequency of flood events significantly influence CH \(_{4}\) production from rice paddies. While some studies have explored the effects of heavy rainfall on flood occurrences, few have examined their direct connection to methane emissions. This study is divided into two parts: first, we use machine learning (ML) techniques to predict patterns of rainfall across India; second, we suggest a framework that uses explainable AI (XAI), such as SHapley Additive exPlanations (SHAP), to obtain the most important features which affect the CH \(_{4}\) emissions from rice paddies. To assess the performance of these models, we used 10-fold cross-validation, which showed that the Multi-Layer Perceptron outperformed the others. Furthermore, this study highlights the relevance of hyperparameter adjustment in enhancing model accuracy and finding significant features, which is very useful in environmental monitoring applications.

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

Leveraging Explainable AI for Rainfall Prediction and Its Impact on Methane Emissions from Rice Paddies Using Multivariate Environmental Data

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ankan Bhattacharya,
  • Sarbani Palit,
  • Fathima Nuzla Ismail,
  • Abira Sengupta

摘要

A direct relationship exists between methane (CH \(_{4}\) ) emissions from rice paddies and their environmental impacts that is already recognised in the scientific community. Changes in rainfall patterns and the increasing frequency of flood events significantly influence CH \(_{4}\) production from rice paddies. While some studies have explored the effects of heavy rainfall on flood occurrences, few have examined their direct connection to methane emissions. This study is divided into two parts: first, we use machine learning (ML) techniques to predict patterns of rainfall across India; second, we suggest a framework that uses explainable AI (XAI), such as SHapley Additive exPlanations (SHAP), to obtain the most important features which affect the CH \(_{4}\) emissions from rice paddies. To assess the performance of these models, we used 10-fold cross-validation, which showed that the Multi-Layer Perceptron outperformed the others. Furthermore, this study highlights the relevance of hyperparameter adjustment in enhancing model accuracy and finding significant features, which is very useful in environmental monitoring applications.