Answers are grounded in this paper and cite their sources.
This paper examines the role of data science in transforming insurance through a bibliometric mapping of machine learning research on motor insurance claims management. Digitalization and technologies such as artificial intelligence, IoT and telematics are reshaping underwriting, claims processing, pricing and fraud detection, and motor insurance is a particularly prominent field of application. The study combines a narrative literature review with a bibliometric analysis of 195 Scopus-indexed documents published between 1994 and 2026, conducted with the Bibliometrix R package and Biblioshiny. The corpus is a moderately sized, interdisciplinary and fragmented field: it grows by 6.27% annually, conference papers (105) outnumber journal articles (68), and about 87% of documents appeared between 2018 and 2024. Keyword co-occurrence and thematic mapping reveal three research communities: automated claims processing based on computer vision and deep learning, fraud detection using ensemble methods, and predictive analytics for pricing, risk assessment and customer-facing operations. Trend analysis shows a shift from a method-centric to an application- and safety-centric orientation, with growing attention to deployment, security and customer trust. Theoretically, the study provides a structured map of the field’s thematic evolution. Practically, it suggests that insurers should prioritize explainability infrastructure and bias supervision alongside predictive accuracy.
No takes yet. Share an insight, caveat, or question.
Stanković et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: