Scholarlink

The escalating problem of plastic pollution has become a critical issue in the environment, and a move towards environmentally friendly consumption behaviour is imperative in this regard. With concern about plastic trash growing, it is important to examine factors that drive consumption behaviour towards choosing goods with less use of plastic. Social norms, perceived behaviour control, values, and awareness about the environment have been proven to have a significant impact on consumption behaviour. The current work utilizes machine learning algorithms to explore these determinants in a pool of 500 respondents. Analytic techniques utilized in the work include Logistic Regression, Random Forest Classifier, and Decision Tree Classifier, with Logistic Regression providing 89% predictive accuracy. For model interpretability, Explainable AI (XAI) and SHAP (Shapley Additive Explanations) have been utilized, providing rich insights about factors driving consumption behaviour in a healthy environment. The current study confirms the effectiveness of combining theoretical frameworks with machine learning algorithms in curbing degradation in the environment and driving consumption behaviour towards a healthy environment.

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

Developing a Framework to Distinguish Green Behaviour Among Plastic Consumers Using Machine Learning Approaches

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sofia Noor Rafa,
  • Tasfia Tahsin Annita,
  • Mohammad Tahmid Noor,
  • Samura Rahman,
  • Victor Dhrubo,
  • Md. Al Ehtesum Korim,
  • Ahmed Wasif Reza

摘要

The escalating problem of plastic pollution has become a critical issue in the environment, and a move towards environmentally friendly consumption behaviour is imperative in this regard. With concern about plastic trash growing, it is important to examine factors that drive consumption behaviour towards choosing goods with less use of plastic. Social norms, perceived behaviour control, values, and awareness about the environment have been proven to have a significant impact on consumption behaviour. The current work utilizes machine learning algorithms to explore these determinants in a pool of 500 respondents. Analytic techniques utilized in the work include Logistic Regression, Random Forest Classifier, and Decision Tree Classifier, with Logistic Regression providing 89% predictive accuracy. For model interpretability, Explainable AI (XAI) and SHAP (Shapley Additive Explanations) have been utilized, providing rich insights about factors driving consumption behaviour in a healthy environment. The current study confirms the effectiveness of combining theoretical frameworks with machine learning algorithms in curbing degradation in the environment and driving consumption behaviour towards a healthy environment.