?? Day79 of #100DaysOfPython ??

Today, we're diving into few other ways to handle missing values via Frequent Categories Imputation!

Frequent Categories Imputation is used to handle missing values by replacing them with the most frequent category.

Checkout more techniques on handling missing values!

Let's dive into an example on imputing missing observations in the feature by replacing missing values with most frequent categories through different methods:

Link to titanic dataset

1. Adding a variable to compute NaN

2. In case the feature has many frequent categories - the NaNs are replaced with a new category (most frequently used technique)

Replacing the missing values with NaN values has an advantage where any impact of the model's predictive power can be determined if it is because of the missing values.