Optimized Machine Learning Models for Poverty Detection: A Scientific Review of Multidimensional Approaches

Abdulrehman Mohamed, Fullgence Mwakondo, Kelvin Tole and Mvurya Mgala
Additional contact information
Abdulrehman Mohamed: Institute of Computing and Informatics, Technical University of Mombasa, Tom Mboya Street Tudor, Mombasa
Fullgence Mwakondo: Institute of Computing and Informatics, Technical University of Mombasa, Tom Mboya Street Tudor, Mombasa
Kelvin Tole: Institute of Computing and Informatics, Technical University of Mombasa, Tom Mboya Street Tudor, Mombasa
Mvurya Mgala: Institute of Computing and Informatics, Technical University of Mombasa, Tom Mboya Street Tudor, Mombasa

International Journal of Research and Scientific Innovation, 2025, vol. 12, issue 3, 1081-1090

Abstract: This paper enhances the discussion on machine learning (ML) models for poverty detection by introducing empirical validation, comparative performance analysis, and practical deployment strategies. We validate the proposed Optimized Machine Learning Model (OMLM) through experiments on real-world datasets. A comparative study against existing poverty detection models, including logistic regression, decision trees, and convolutional neural networks (CNNs), highlights OMLM’s superior adaptability and accuracy. The paper further explores data limitations, computational efficiency, and regional performance variations. Finally, a novel optimization technique, combining Genetic Algorithms (GA) with Reinforcement Learning (RL), is introduced to refine predictive accuracy and real-time adaptability. Practical implementation details, including data processing pipelines, cloud-based deployment, and integration into governmental policy frameworks, are discussed to enhance the model’s real-world applicability. This study contributes to advancing ML applications in poverty detection, reinforcing its role in data-driven policymaking and targeted socio-economic interventions.

Date: 2025
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.rsisinternational.org/journals/ijrsi/d ... ssue-3/1081-1090.pdf (application/pdf)
https://rsisinternational.org/journals/ijrsi/artic ... ensional-approaches/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bjc:journl:v:12:y:2025:i:3:p:1081-1090

Access Statistics for this article

International Journal of Research and Scientific Innovation is currently edited by Dr. Renu Malsaria

More articles in International Journal of Research and Scientific Innovation from International Journal of Research and Scientific Innovation (IJRSI)
Bibliographic data for series maintained by Dr. Renu Malsaria ().