Literature survey on DEA in the insurance industry with a focus on identification of research hotspots with text mining

Abstract

DEA is a frequently used non-parametric methodology for measuring the relative efficiency of Decision-Making Units (DMUs) that use the same inputs to produce the same outputs. Emrouznejad and Yang (2018) provided a literature survey on DEA with 10,300 peer-reviewed journal articles from 1978 to the end of 2016. Our article focuses on DEA applications in the insurance industry in convergence with the existing relevant literature as Kaffash et al (2020), who have surveyed 132 DEA articles in the insurance industry for the period from 1993 to 2018. We include particular keyword analyses necessary to identify research hotspots in different periods. This article aims to conduct a bibliometric analysis of DEA-published documents (articles in journals and book chapters) in the insurance industry from 1993 to 2021, focusing on identifying research hotspots based on keyword co-occurrence analysis. We have analyzed published documents from relevant databases, such as Scopus, Web of Science, Ebsco and ProQuest. We use descriptive analytics and text mining as the main methods in our analysis. We provide descriptive statistics for articles per year and category of the insurance industry, geographical distribution, top five journals and authors by citations, and citation analysis. An additional qualitative factor of our article is in-depth keyword co-occurrence analysis by using text mining to identify research hotspots in the insurance industry. Our analysis aims to contribute to researchers and insurance practitioners as an empirical and applicative point for initiating and developing research.