Data Exploration

The single most important thing for a beginner to know about machine learning (ML) is that machine learning is not magic. Taking a large dataset and naively applying a neural network to it will not automatically give you earth-shaking insights. ML is built on top of sound and familiar mathematical principles, such as probability, statistics, linear algebra, and vector calculus—voodoo not included (though some readers may liken vector calculus to voodoo)!

We will be covering the following topics in this chapter:

• An overview
• Variable identification
• Cleaning of data
• Transformation
• Types of analysis
• Missing values treatment
• Outlier treatment