How to detect and deal with Outliers

Outlier replacement strategy:
Replace outliers with missing values

Outlier replacement strategy: Remove outlier rows

Outlier replacement strategy: Remove non-outlier rows

How to detect and deal with Outliers

Starts with the Iris dataset, then applies two main strategies for handling outliers: IQR-based outlier detection (to either replace or remove outliers) and Z-score normalization (to standardize numeric columns). After normalization, it identifies rows where the absolute z-score is greater than 2, marking them as outliers, and finally filters out these outlier rows to keep only the typical data points for further analysis.