Mapping the Evolution of TinyML: A Bibliometric and Science-Mapping Analysis of Global Research

Context: Tiny Machine Learning (TinyML) has emerged as an important edge-intelligence paradigm for deploying machine-learning models on resource-constrained and low-power devices. Its rapid expansion across diverse application domains has created a need for systematic mapping of the research landscape.

Aim: This study maps the development, scholarly impact, collaboration patterns, and thematic evolution of global TinyML research.

Approach: Bibliographic records were retrieved from Scopus using “TinyML” and “Tiny Machine Learning.” Following systematic screening and data cleaning, 1,291 publications covering 2020–2026 were analyzed using bibliometric and science-mapping techniques.

Key Findings: TinyML research exhibited rapid growth, with an annual growth rate of 69.84% and a peak complete-year output of 422 publications in 2025. India, China, and Italy were the leading corresponding-author countries, while the University of Bologna was the most productive institution. Keyword analysis revealed a thematic transition toward learning systems, edge computing, embedded systems, and real-time applications.

Interpretation: The findings demonstrate that TinyML has developed into a rapidly expanding, internationally collaborative, and increasingly application-oriented research domain.

Contribution: This study provides an updated quantitative mapping of TinyML scholarship and identifies influential contributors, collaboration patterns, established themes, and emerging research directions.