
Chapter 9
Novelty Detection in Data Streams
Novelty detection is a learning task that consists of the identification of new
or unknown concepts that the learning system is not aware of during training.
This is one of the fundamental requirements of a good classification or iden-
tification system, since sometimes the test data contain information about
concepts that were not known at the time of training the model. In time-
dependent applications, novelty detection represents an important challenge,
since concepts are hardly ever constant. This chapter surveys the major ap-
proaches for the detection of novelty in data streams.
1
9.1 Introduction
Novelty