In this chapter, we saw one of the simplest techniques of machine learning called k-nearest neighbors. We also looked at an example of KNN which predicts the rating for a movie. We analysed the concepts of dimensionality reduction and principal component analysis and saw an example of PCA, which reduced 4D data to two dimensions while still preserving its variance.
Next, we learned the concept of data warehousing and saw how using the ELT process instead of ETL makes more sense today. We walked through the concept of reinforcement learning and saw how it is used behind the Pac-Man game. Finally, we saw some fancy words used for reinforcement learning (Q-learning, Markov decision process, and dynamic learning). In the next chapter, we'll see how to deal with real-world data.
