Answers are grounded in this paper and cite their sources.
Exploring data mining methods and algorithms through reviews is crucial, as it might be useful for researchers who need effective decision-making criteria in applying relevant methods and algorithms for specific data mining tasks. So far, most of the previous reviews still lack an in-depth discussion of what data mining practices offer regarding trends and opportunities for future research direction. Therefore, we conducted a review to systematically assess the current practice of data mining methods and algorithms in public health research and identify future research direction opportunities. The review follows PRISMA technique, examining 40 peer-reviewed articles published in computer science proceedings and journals between 2016 and 2025. The review resulted in three key contributions. 1) Exploration of four data mining methods across nine public health research areas by trends. 2) Discovery of 15 data mining algorithms categorized by data mining methods. 3) Mapping of data mining methods and algorithms within the public health research area for future research direction opportunities. The review examined popular and underutilized data mining methods and algorithms, highlighting the need for further investigation into optimizing the underutilized algorithms to enhance computational efficiency against outliers and noise. We suggest planning hybrid methods that integrate the strengths of well-established algorithms and computational techniques to improve reliability and accuracy.
No takes yet. Share an insight, caveat, or question.
Alebaba et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: