Welcome to the "Understanding the Causes of Depression" project. Depression is a complex mental health condition, and its causes can vary widely from person to person. In this project, we aim to delve into the factors that most significantly influence an individual's likelihood of experiencing depression. Through the application of various machine learning models on provided datasets, we seek to uncover patterns based on the available features. The output of our model training will highlight the features that have the greatest impact on depression, shedding light on potential causes.
Depression is a major public health concern, affecting millions of individuals worldwide. Identifying the key factors contributing to depression can greatly enhance our understanding of this condition. We harness the power of several machine-learning models to explore the factors behind depression. The models employed in this project include:
- XGBoost
- LightGBM
- K-Nearest Neighbors (KNeighbor)
- Logistic Regression
To ensure that our machine learning models perform optimally, we utilize Optuna for hyperparameter tuning. Optuna allows us to fine-tune the models, optimizing their performance and enhancing their predictive accuracy.
The interactive EDA dashboard and machine learning model results can be accessed here : Link to dashboard
The final notebook : notebook
The data consists of a study about the life conditions of people who live in rural zones. Here to get the dataset.
Understanding the causes of depression is a multifaceted endeavor, and this project represents a crucial step toward shedding light on this intricate issue. By identifying the factors that most strongly influence depression, we can advance our knowledge and ultimately work towards more effective prevention and support strategies. Thank you for your interest in our research.