Handling Mixed Variables in Feature Engineering: A Practical Guide with Code

Author(s): Souradip Pal

Originally published on Towards AI.

This member-only story is on us. Upgrade to access all of Medium.

A girl looking at a screen containing mixed variables. Source: Image generated by Dall-E

Imagine you’re working on a brand-new data project, the kind that makes your hands twitch with excitement. Everything seems perfect, and then you hit a roadblock: mixed variables. Yep, those quirky features that contain both numbers and characters. Whether it’s a cell with something like “A1” or a column that stubbornly holds both strings and integers, mixed variables can be a real headache.

But don’t worry! In this blog, we’ll dive into two common scenarios you might face when dealing with mixed data and walk through how to fix them with practical code examples. Ready? Let’s get started!

Picture this: You’re analyzing a dataset for a retail store. One of the columns contains stock codes — like “A1”, “B3”, and “C7”. These codes aren’t just random; they contain valuable information, like a product type (“A”, “B”, “C”) and a version number (1, 3, 7). But because they’re squished together in one cell, your machine-learning model will probably throw a fit if you try to use them directly. What to do?

You need to split these values into their meaningful… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI