Analysis of research structure and scientific trends in artificial intelligence, machine learning, and deep learning in football: a bibliometric approach

Abstract

Purpose The integration of artificial intelligence (AI), machine learning (ML), and deep learning (DL) into football research has grown rapidly, creating new opportunities in tactical analysis, performance optimization, injury prediction, and talent identification. Despite this progress, comprehensive overviews of the field remain limited. This study provides a systematic bibliometric analysis of 1,112 peer-reviewed articles published between 2010 and 2024 in the Scopus database, aiming to map the intellectual structure, identify thematic clusters, and explore collaboration trends in AI-driven football research. Design/methodology/approach Articles and reviews in English were retrieved from Scopus using the terms “artificial intelligence,” “machine learning,” and “deep learning” combined with “football” and “soccer.” After data cleaning and normalization, analyses were conducted with the R Bibliometrix package and VOSviewer. The study assessed publication and citation patterns, co-authorship networks, keyword co-occurrence, co-citation, bibliographic coupling, and thematic evolution. Findings The results show a sharp increase in publications since 2017, reflecting the growing role of AI in football. China leads in the number of articles, while Germany and Australia demonstrate higher citation impact. International collaboration accounts for 18.22% of studies, emphasizing the global and interdisciplinary nature of this field. Research clusters highlight key areas such as match prediction, injury forecasting, tactical decision-making, computer vision, reinforcement learning, sentiment analysis, and simulation modeling. Keyword analysis indicates the prominence of machine learning, deep learning, and performance analytics, with increasing attention to robotics, sports medicine, and real-time data processing. Originality/value This study presents the first systematic bibliometric mapping of AI, ML, and DL in football using Scopus data up to June 2024. It offers a comprehensive understanding of research productivity, collaboration, and knowledge gaps, providing strategic directions for advancing AI applications in football analytics.

Ähnliche Arbeiten