Hands-On Data Science for Marketing

Challenges in choosing the right model

Choosing a machine learning algorithm for a given data science project is more difficult than it sounds. Some algorithms work more like a black box, where you do not know how an algorithm makes predictions or decisions. For example, it is quite difficult to understand how a trained random forest model makes predictions on the output from the input. The decisions are made from hundreds of different decision trees, where each tree works differently with different decision-making criteria, and this makes it difficult for a data scientist to fully understand what happens in between the input and the output.

On the other hand, linear models, such as logistic regression models, tell us exactly how they are ...