Suicide: An Analysis and Prediction Using Pyspark and Machine Learning Models

The World Health Organization (WHO) reported in 2021 that more over 700,000 people had committed suicide. Suicide can be stopped, but most efforts have so far been ineffective. However, the application of machine learning presents fresh chances to improve prediction accuracy and advance the cause of suicide prevention. This study uses data-driven methodologies to analyze and predict suicide, and presents the same in this report. Machine learning algorithms and PySpark programs are used to analyze a dataset of demographic, socioeconomic, and psychological characteristics related to suicide instances from a Kaggle dataset comprising of 237519 rows and 7 columns. The study creates prediction models like linear regression, decision tree, decision tree regressor and bagging regressor. The results support efforts to prevent suicide by assisting in the prioritization of resources and focused interventions.