50% of the dataset is extracted from Kaggle, 25% is extracted from the UCI machine learning repository, and the remaining 25% of the data is extracted from the IBM data scientist page. All the data are combined into a single dataset and have been made available for open-AI projects.