Machine learning in Social Sciences and Humanities research: A structured literature review

Abstract

This research explores the growing role of machine learning in social sciences and humanities research, aiming to provide a comprehensive bibliometric review of its applications, trends, and impact. A systematic approach was employed, combining bibliometric analysis and visualization techniques to analyze a dataset extracted from the Scopus database, focusing on scholarly works related to machine learning in disciplines such as sociology, philosophy, history, and economics. The analysis highlights the increasing volume of publications and the evolving research landscape, with notable growth in recent years, particularly after 2005. Key findings indicate that machine learning is primarily applied in areas such as decision-making, sustainability, social media analysis, and natural language processing, reflecting its diverse potential in addressing complex societal issues. The study also reveals a strong collaborative nature, with a significant percentage of publications involving international co-authorship. Geographic analysis shows that countries like the United Kingdom, India, and the United States are leading contributors, while emerging nations also make noteworthy contributions. The findings suggest that machine learning is becoming an essential tool for social and cultural research, offering new insights into behavioral patterns, governance, and global challenges. In conclusion, the interdisciplinary nature of this approach is emphasized, along with the need for continued methodological innovation and cross-border collaboration to enhance its impact in these fields.