An Apriori Algorithm-Based Association Rule Analysis to detect Human Suicidal Behaviour

Abstract

Suicide is a major cause of death. It is also a complex public health issue and often preventable with timely intervention. Overall, the rate of suicide is increasing for various reasons. In our study, we use an association rule analysis to find the most important rules to predict suicidal behavior from an available data set. One of the most powerful machine learning algorithms available for identifying associations within databases is the Apriori algorithm. We used this algorithm to analyze association rules of suicidal behavior using a dataset of 1250 instances and 27 impactful features. These include daily activities, family background, and answers to mental questionnaires and have been analyzed to find combinations that are associated with suicidal behavior. The study has resulted in some key rules for human suicidal behavior. The Apriori method has been used to identify the eight most significant rules with the support of 0.25 and the confidence of 0.90.