Leveraging Machine Learning to Improve Social Justice in the Digital Age

Abstract

The development of digital technology has encouraged the use of machine learning in various sectors to improve the efficiency and accuracy of decision-making. This study aims to analyze the role of machine learning in improving social justice in the digital era through a multidisciplinary approach that integrates technological, social, and policy aspects. The study used the Systematic Literature Review (SLR) method with reference to PRISMA guidelines, with data sources derived from reputable scientific databases for the period 2020–2025. A total of 30 selected articles were analyzed using a thematic analysis approach to identify key patterns and relationships between variables. The results show that machine learning can increase service efficiency by 80% and predictive accuracy by 73% in various sectors such as health, education, and social services. In addition, this technology contributes to increasing service access (70%) and expanding digital inclusion (65%). However, this study also identified key challenges in the form of algorithmic bias (75%), the digital divide (60%), and low technological literacy. From a policy perspective, regulations based on the principles of transparency, accountability, and fairness are important factors in controlling the risks of machine learning implementation. This research produces a multidisciplinary conceptual model demonstrating that social justice in digital societies is the result of the interaction between technology, social factors, and policy. This research provides theoretical contributions to the development of the concept of digital equity and offers practical implications for policymakers and technology developers in designing more equitable, transparent, and inclusive machine learning systems.