Explain common techniques for transforming categorical features into numerical formats for machine learning.
Task: Describe common techniques for transforming categorical features in a dataset into a numerical format suitable for machine learning models. Context: You have a dataset containing one or more categorical columns (e.g., 'color', 'city', 'product_type') and need to prepare them for an algorithm that requires numerical input. Techniques to cover: 1. One-hot encoding 2. Label encoding 3. Ordinal encoding 4. Target encoding (briefly explain concept) For each technique, explain: - When to use it. - Its advantages and disadvantages. - A simple conceptual example. Output Format: Provide a clear explanation for each technique in a bulleted list.
Generate ideas for creating new numerical features from existing ones in a dataset, focusing on simple transformations.
Generate a simple strategy to identify and handle missing values in a given dataset using basic methods.
Brainstorm simple ideas for creating new, useful features from existing numerical or categorical columns in your dataset to enhance model performance.
