Scholarlink

This paper presents an adaptive decision support framework utilizing Human-Centric Artificial Intelligence (HCAI) to aid poverty reduction efforts among marginalized communities. The model dynamically incorporates real-time data inputs such as income distribution, welfare metrics, and poverty headcount ratios. By integrating community feedback loops, the system continuously adjusts to the needs of local populations, providing contextually relevant insights and tailored policy recommendations. Leveraging machine learning algorithms, including regression, this approach achieves high predictive accuracy and actionable, region-specific insights for decision-makers. Findings suggest the model’s potential for scalable, ethically aligned poverty alleviation. Different regression model techniques were applied to the model. XGBoost outperforms other models, achieving an accuracy of 63% and the least error of 12% so it can be considered the optimal model. The residual plots have confirmed the performance of the XGBoost model for capturing the complex relationship of the data more effectively.

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

An Adaptive Framework for Empowering Marginalized Communities Through Human-Centric Artificial Intelligence (HCAI) for Decision-Making in Poverty Reduction Using Regression Techniques

  • 章节

  • 发表日期：

    2025年

查看全文
  • R. Vidya,
  • P. Deepan,
  • N. Arul

摘要

This paper presents an adaptive decision support framework utilizing Human-Centric Artificial Intelligence (HCAI) to aid poverty reduction efforts among marginalized communities. The model dynamically incorporates real-time data inputs such as income distribution, welfare metrics, and poverty headcount ratios. By integrating community feedback loops, the system continuously adjusts to the needs of local populations, providing contextually relevant insights and tailored policy recommendations. Leveraging machine learning algorithms, including regression, this approach achieves high predictive accuracy and actionable, region-specific insights for decision-makers. Findings suggest the model’s potential for scalable, ethically aligned poverty alleviation. Different regression model techniques were applied to the model. XGBoost outperforms other models, achieving an accuracy of 63% and the least error of 12% so it can be considered the optimal model. The residual plots have confirmed the performance of the XGBoost model for capturing the complex relationship of the data more effectively.