Predict-Clinical-Insomnia-Among-Cancer-Survivors-Using-Machine-Learning-Approaches

Main tasks:

1. Process baseline data of YOCAS II randomized controlled trial among 740 cancer survivors;
2. Develop Machine Learning models to predict ISI (Insomnia severity index) and classify severe Insomnia (ISI >= 19);
3. Provide clinically meaningful interpretation.

Descriptive analysis: 1: ISI distribution 2: Predictors 3: Correlation analysis 4: Patient-symptom network

Machine Learning models: 1: Data preprocessing 2: Feature selection 3: Regression models 4: Classification models

Note: All data merging, data preprocessing, data imputations, features engineering, model training, tuning, and selection is included in the Data_exploration.ipynb. Patient-symptom network and symptoms clustering is included in Symptoms Clustering .ipynb.