Electrocution is one of the main causes of workplace deaths in the construction industry. This paper presents a framework for identifying electrocution risk factors and exploring the correlations between them, with the aim of assisting accident prevention research. Specifically, the Haddon Matrix is used to extract the risk factors from 193 investigation reports of electrical shock accidents from 2012-2019, and the Apriori algorithm is applied to examine the potential relationships between these factors. Based on association rules using three criteria: support ( S), confidence ( C) and lift ( L), the betweenness centrality is then introduced to optimize association rules and find the most important rules though comparison. The results show that after optimization, some of these critical rules rise significantly in rank, such as Workplace: indoor → No CPR provided. Through these ranking changes, the focus of safety management is clarified, and finally, based on a comprehensive analysis of association rules, targeted accident prevention measures are suggested.