Implemented Machine Learning Models to PredictLife Satisfaction Using Demographic and Religious Data

The study aims to demonstrate the effectiveness of machine learning in predicting life satisfaction based on demographic and religious factors using data from the World Values Survey (Wave 7). Key variables such as age, sex, income, education level, and religious practices were analyzed using classification models—Random Forest, XGBoost, and Decision Tree. XGBoost emerged as the most effective model. The user interface developed from this model provides predictions of life satisfaction and personalized recommendations, potentially guiding targeted mental health interventions and resource allocation. While the study offers valuable insights, future research could refine the model by incorporating additional psychosocial variables and testing deep learning techniques, such as recurrent neural networks (RNNs). Integrating realtime or longitudinal data could also improve the model's adaptability across diverse populations, advancing its applicability in community-based programs and policymaking.