Summary

Between decision trees, Naïve Bayes classification, feature extraction, and K-means clustering, we have seen that machine learning goes way beyond the simplicity of linear and logistic regression and can solve many types of complicated problems.

We also saw examples of both supervised and unsupervised learning and in doing so became familiar with many types of data science related problems.

In the next chapter, we will be looking at even more complicated learning algorithms including artificial neural networks, and ensembling techniques. We will also see and understand more complicated concepts in data science, including the bias-variance tradeoff, as well as the concept of overfitting.