St Edward's Academic Review 2025

7

To what extent is data mining effective in predicting natural disasters? By Carissa Moses-Saromi

Introduction

Data science, an emerging field that combines Mathematics, Artificial Intelligence, machine learning and statistics, is a collection of principles, problem definitions, algorithms, and processes for discovering and extracting non-obvious and useful patterns from large data sets (Kelleher & Tierney, 2018). There are various applications of data science, including: in healthcare to predict disease; in sports to evaluate the performance of athletes; in e-commerce to automate digital advertisement placements; in gaming to improve online gaming experiences; in financial technology (Fintech) to assist with the creation of financial profiles and predictive models based on historical payroll data among others (Rice, Powers, & Oppermann, 2023). One of the major applications of data science is in predicting extreme weather phenomena using historical data and more recent trends.

historical data as input to predict new output values, machine learning enables software applications to become more accurate at predicting outcomes without being explicitly programmed to do so (Burns, 2021). Machine learning focuses on designing and evaluating algorithms that can be used to extract patterns from data (Kelleher & Tierney, 2018). Kelleher and Tierney also outlined that data mining deals with the analysis of structured data and often implies an emphasis on commercial applications. This essay will examine the use or non use of data science and machine learning techniques during the Haiti earthquake in 2010 and the Turkey-Syria earthquake in 2023 to determine how effective it has become in predicting natural disasters. From my findings so far, data science and machine learning techniques have not been used to predict natural disasters even though research has been conducted and structures put in place for this purpose.

This essay is aimed at explaining the impact of data science on predicting natural disasters and how the field has evolved over the last decade. Martin Degg in 1992 formulated a definition of natural disasters as events that result from natural processes that cannot be predicted and are, therefore, difficult to avoid. This directly relates to data science because it can facilitate the extraction of different types of patterns. For example, it identifies groups of items that exhibit similar behaviour and tastes (clustering), identifies products that are purchased together in businesses, extracts patterns that identify strange or abnormal events (anomaly/outlier detection), and identifies patterns that aid in classification (prediction). Many aspects of data science have emerged in other related fields such as machine learning and data mining. The terms data science, machine learning and data mining are often used interchangeably (Burns, 2021). Each of these fields is concerned with improving decision-making through data analysis. For instance, by using