A Bibliometric and Content Analysis of Applied Machine Learning Research in Social Media Marketing
İnsan ve Toplum Bilimleri Araştırmaları Dergisi · 30.06.2026 · 15 (2) · 1423-1445
DOI: 10.15869/itobiad.1855614 ↗Kurz gefasst
Fachartikel in der Zeitschrift İnsan ve Toplum Bilimleri Araştırmaları Dergisi, online erschienen am 30.06.2026. Verfasst von Hakan Çanlı und 1 weiteren Autor:innen. Die Arbeit nennt 67 Literaturangaben. Gefördert von: The author(s) acknowledge that they received no external funding in support of this research..
Abstract (Originalsprache)
This study analyses 632 articles from Web of Science and SCOPUS databases at the intersection of machine learning, social media and marketing disciplines. Firstly, quantitative content analysis was performed based on four coding schemas: types of machine learning methodologies, social media platforms, contributions to marketing and research outputs. Secondly, bibliometric analysis identified scientific production trends leveraging these classifications. Finally, network mapping techniques were used to visualize collaboration in the field. Production trends reveals that the field is vibrant and growing exponentially with a doubling time of 2.2 years. Content analysis exposes the gaps and critical imbalances in the literature. While use of traditional machine learning and deep learning methods dominate cumulatively, transformer models have overtaken traditional deep learning in annual usage. Thus, the research field is currently undergoing a methodological shift to transformer and large-scale foundation models. From marketing point of view the field mainly focuses on producing customer insight and marketing research (46%) with limited contribution to other marketing tasks. Microblogging platforms (43%) are positioned as the dominant data source in the field and functions as a laboratory for academic research. The ratio between research focusing on methodological development (49%) and marketing insight (51%) suggests the field is still building its tools. The network analysis reveals centralized country-level but decentralized institutional and author-level collaboration, highlighting that the domain is fragmented. To address these gaps and imbalances in the domain, it is recommended that future research apply advanced models for relatively undersupported marketing functions. This study utilizes a tailored, scalable, prompt engineering supported, and replicable methodology which can be utilized for future bibliometric analysis.
Förderung
- The author(s) acknowledge that they received no external funding in support of this research.
Zitieren
APA
Çanlı, H., & Atalık, Ö. (2026). A Bibliometric and Content Analysis of Applied Machine Learning Research in Social Media Marketing. İnsan ve Toplum Bilimleri Araştırmaları Dergisi, 15(2), 1423-1445. https://doi.org/10.15869/itobiad.1855614
Vancouver
Çanlı H, Atalık Ö. A Bibliometric and Content Analysis of Applied Machine Learning Research in Social Media Marketing. İnsan ve Toplum Bilimleri Araştırmaları Dergisi. 2026;15(2):1423-1445. doi:10.15869/itobiad.1855614
BibTeX
@article{anl2026,
author = {Çanlı, Hakan and Atalık, Özlem},
title = {A Bibliometric and Content Analysis of Applied Machine Learning Research in Social Media Marketing},
journal = {İnsan ve Toplum Bilimleri Araştırmaları Dergisi},
year = {2026},
volume = {15},
number = {2},
pages = {1423--1445},
doi = {10.15869/itobiad.1855614}
}
EUVIMED