The purpose of this notebook is to analyze the data and then modelize the mortality based on the features given in the dataset.
To reach this goal, we will follow the following steps:
produce an exploratory analysis of the data, the role of which is to assess the quality of the data and especially to identify highly correlated features build a model based on stacking optimize this model in terms of the number of basic models and the number of features. For this experience, we will keep:
5 models having a train score greater than 0.8 5 features having a correlation coefficient smaller than 0.75. The results are stored in a database in Kaggle, and they are analyzed in this notebook.