this project contains a full knowledge discovery path on stroke prediction dataset. list of steps in this path are as below:
- exploratory data analysis available in P2.ipynb
- data preprocessing (takeing care of missing data, outliers, etc.) available in preparation.ipynb
- preparing two new datasets of cleaned data, one for people 18 to 90 years old and one for 18 to 75 years old available in preparation.ipynb
- training and testing models available in models.ipynb
- evaluation and error analysis available in models.ipynb
we hope to help people in danger of brain stroke, so far based on this dataset we can inform 83% of stroke victims beforehand.
dataset link:https://www.kaggle.com/fedesoriano/stroke-prediction-dataset