K-means and hierarchical based clustering in suicide analysis

Abstract

Machine learning is the intriguing area of research that spreads across all domains helping in providing quality decisions. Demographic have more influence in social happenings along with various personal and social factors. Suicide analysis is one such issue to be handled with great concern that will provide precautionary based on situations. Suicide prediction can be carried on using data mining that can be used to predict the suicide earlier so that it can be prevented. Suicide is an action resulting in death performed by themselves. Common factors that influence the rate of suicides are cause, method of suicide, year, gender, educational qualification, social status. For this clustering technique in datamining that falls under unsupervised provides great platform. Silhouette score is used for mapping the number of cluster to get the good clustering. Various plots like box plot, scatter plot and so on helps to provide greater insight. Based on analysis the required remedial could be arrived.