Emoji and Text Sentiment Classification: A Scientometric Analysis

The integration of emojis with textual content in digital communication has transformed sentiment analysis, necessitating advanced methodologies to decode nuanced emotions in hybrid data. This study presents a comprehensive scientometric analysis of 487 publications (2008–2024) on emoji–text sentiment classification, with 43.9% of research rooted in computer science. Data were systematically retrieved from IEEE, Scopus, Web of Science, and EBSCO using predefined search queries, followed by rigorous analysis employing logistic growth modeling, social network analysis, and citation mapping. Key findings reveal an accelerated growth phase (147.83% peak in 2017–2018) and subsequent stabilization by 2022–2023. Collaboration networks exhibit strong regional clustering, while thematic evolution progressed from basic sentiment analysis to specialized domains like emotion detection and socio-cultural implications of digital communication. The study identifies influential authors, institutions, and journals, with Lecture Notes in Computer Science emerging as the most impactful source (h-index: 7, 122 citations). Methodologically, the integration of VOSviewer for network visualization and R-based Scientometric tools provided granular insights into co-citation patterns and intellectual structures. These results establish a strategic framework for future research, emphasizing the interdisciplinary convergence of computational techniques and socio-linguistic studies in emoji–text sentiment analysis.