A review of machine learning for big data analytics: bibliometric approach

ABSTRACT

The amalgamation of machine learning and big data has led to a revolution in data science with several influencing applications to various domains. To gain insights on the current research trends on machine learning for big data analytics, this study follows a bibliometric analysis methodology of citation data to review and quantitatively assess the explosion and impact of literature and research performance in this vibrant research area, which has witnessed rapid changes and rising interest in business, industry and academia. Using a variety of bibliometric measures and visualisation techniques, the paper examines and identifies several related issues including research productivity and directions, major contributors, publication trends and growth rates, citation and collaboration analysis, and others. The relevant bibliographic units for the study were collected from the Core Collection of the Web of Science bibliographic database. Nearly all the relevant publications prior to February 2018 were included in the analysis. The overwhelming productivity and wide-spread applications in several multidisciplinary domains have been revealed, with one-to-two ratio of journal to conference publications. Three countries (USA, China, India) are dominating the research output with more than two-thirds of the total productivity.