Data mining and machine learning techniques applied to public health problems: A bibliometric analysis from 2009 to 2018

The objective of this paper is to present a bibliometric analysis of the applications of Data Mining (DM) and Machine Learning (ML) techniques in the context of public health from 2009 to 2018. A systematic review of the literature was conducted considering three major scientific databases: Scopus, Web of Science and Science Direct. This enabled an analysis of the number of papers by journal, the countries where the applications were carried out, which databases are more commonly used, the most studied topics in public health, and the techniques, programming languages and software applications most frequently used by researchers. Our results showed a slight increase in the number of papers published in 2014 and a significative increase since 2017, focusing mostly on infectious, parasitic and communicable diseases, chronic diseases and risk factors for chronic diseases. The Journal of Medical Internet Research and PLoS ONE published the highest number of papers. Support Vector Machines (SVM) were the most common technique, while R and WEKA were the most common programming language and software application, respectively. The U.S. was the most common country where the studies were conducted. In addition, Twitter was the most frequently used source of data by researchers. Hence, this paper provides an overview of the literature on DM and ML in the field of public health and serves as a starting point for beginner and experienced researchers interested in this topic.