Optimizing the Classification of Social Media's Impact on Mental Health Using Random Forest With Gridsearchcv Hyperparameter Tuning

The rapid growth of social media has significantly impacted various aspects of life, including individual mental health. This study aims to optimize a classification model for identifying the impact of social media usage on mental health using the Random Forest algorithm. The dataset incorporates various indicators of social media usage behavior and the psychological conditions of respondents. The research process encompasses data preprocessing, model training, and hyperparameter optimization via the GridSearchCV method to enhance model performance. The results demonstrate that hyperparameter tuning improves accuracy, precision, recall, and F1-score compared to the baseline Random Forest model. The resulting optimal model is expected to assist in the early detection of the negative impacts of social media on mental health, thereby serving as a basis for decision-making in the fields of education, healthcare, and public policy.