Summary
In this chapter, we applied our knowledge of grouping statistics to three of the most used Kaggle datasets. We learned that credit card transaction fraud could be analyzed by amounts that are out of the scope of regular transaction money payments. The K-means grouping algorithm can be applied to find out what the quantity and duration of the packets in login attempts are to find out whether there is suspicious login activity or not. Finally, the K-means function can help to conclude that age is not an important factor for money complaints. It is more important to keep the BMI under low-risk levels than the age of an insurance company's clients.
In the next chapter, we are going to build a prediction model based on the possible relationship between two variables.