This presentation is about a master thesis work, a collaboration between a group of researchers from Halmstad University and Statistikkonsulterna AB. The project investigated how machine learning and survival analysis can be used to predict suicidal behavior among older adults in Sweden. The study analyzed historical data from the Swedish National Registry, including medication use, medical conditions, and sociodemographic details. The research aimed to improve prediction accuracy by combining survival analysis techniques — such as Cox Proportional Hazards (CoxPH), Random Survival Forest (RSF), and Gradient Boosting Survival Analysis (GBSA) — alongside sequential machine learning approaches, like Long Short-Term Memory (LSTM) networks. The findings highlighted the potential of utilizing historical data to predict suicide risks using advanced machine learning models.