This project implements a research paper which aims to find a classifier model to predict Mental Health status of individuals. The paper is selected after reviewing various papers on the topic of mental health. The project is done as a course project for the 'BITS F464 Machine Learning' course instructed by Dr. Paresh Saxena at BITS-Pilani Hyderabad Campus.
More details of the project can be found in the report and demo presentation.

The dataset is based on a survey conducted by Busara Center in Western Kenya. It contains information pertaining to individuals' health, financial conditions among other things.
The dataset folder contains training data, testing data and the descriptions of each feature of the dataset.

• Filling NaN values with the average of given data
• Feature selection using Random forest importance technique

The models used are classifiers- SVM, Logistic Regression, Decision Tree, Random Forest, Gradient Boosting and Vote Ensembling of these models. These models are used while tuning various hyper-parameters where necessary.
Finally, the conclusion is that the classifiers- Vote Ensembling, Random Forest and SVM with radial basis kernel yield the best F1 score and accuracy.

• Shashivardhan Vadyala
• Chaitanya Chakka
• K S V L Druthi
• Vansh Madan
• Sheethal Reddy
• Pranay Kumar Dasoju