Outlier treatment is an essential step in data wrangling, which involves detecting and handling data values that are far away from the general values in a dataset. Outliers can cause statistical or machine learning models to perform poorly, as they do not fall within the normal range of values for that attribute. The purpose of outlier treatment is to improve the quality of the data by removing or modifying the outliers, which can help to maintain the size and representativeness of the dataset.

There are several methods of handling outliers in a dataset, including ignoring them, removing them, imputing them, and applying log transformation. Statistical detection is a widely followed method for removing and modifying the outliers using techniques such as trimming, Winsorization, and robust model fitting.

In summary, outlier treatment is an important step in data wrangling that involves detecting and handling data values that are far away from the general values in a dataset. The purpose of outlier treatment is to improve the quality of the data by removing or modifying the outliers, which can help to maintain the size and representativeness of the dataset.