Summary

In this chapter, we understood the logic behind multiclass classification problems. We created a multiclass classifier to predict the most suitable channel to be used to target customers. Through different examples and exercises, we tackled imbalanced datasets. This chapter also gave us an idea of how using different sampling methods can be useful in tackling imbalanced data.

In this book, we have covered several topics that are fundamental to marketing analytics. Beginning with data manipulation and visualization in Python, we covered customer segmentation using unsupervised methods such as clustering, predicted customer spend, and developed ideas for both regression and classification problems using a variety of use cases. Finally, we evaluated and tuned different machine learning models and learned how to handle imbalanced datasets. Following these chapters, you should now be able to think like a data scientist and apply these skills to different marketing scenarios.