Beginning Anomaly Detection Using

 Context-Based  Anomalies

Context-based anomalies consist of data points that might seem normal at first, but 

are considered anomalies in their respective contexts. For example, you might expect 

a sudden surge in purchases near certain holidays, but these purchases could seem 

out of place in the middle of August. As you saw in the example earlier, the person who 

made a high volume of purchases towards Black Friday was not flagged because it is 

typical for people to do so around that time. However, if the purchases were made in a 

month where it is out of place given previous purchase history, it would be flagged as 

an anomaly. This might seem similar to the example brought up for data point-based 

anomalies; the distinction here is that the individual purchase does not have to be 

expensive. If your person never buys gasoline because he/she owns an electric car, 

sudden purchases of gasoline would be out of place given the context. Buying gasoline is 

quite a normal thing to do for everyone, but in this context, it is an anomaly.

Chapter 1   What Is anomaly DeteCtIon?

---

17