Beginning Anomaly Detection Using

 The Three Styles of Anomaly Detection

It is important to note that there are three overarching “styles” of anomaly detection. 

They are

• 

Supervised anomaly detection

• 

Semi-supervised anomaly detection

• 

Unsupervised anomaly detection

Supervised anomaly detection is a technique in which the training data has labels 

for both anomalies and for normal data points. Basically, you tell the model during the 

training process if a data point is an anomaly or not. Unfortunately, this isn’t the most 

practical method of training, especially because the entire data set needs to be processed 

and each data point needs to be labeled. Since supervised anomaly detection is basically 

a type of binary classification task, meaning the job of the model is to categorize data 

under one of two labels, any classification model can be used for the task, though not 

every model can attain a high level of performance. An example of this can be seen in 

Chapter 

7

 with the temporal convolutional network.