Data streams are continuously arriving data in applications such as finance, networks, and sensors. These data streams have to be analyzed so that nuggets can be extracted to determine network intrusions, financial stock market price, and suspicious event detection. As discussed in Chapter 8, there are many challenges that need to be addressed for analyzing data streams. These include infinite length, concept drift, concept evolution, and limited labeled data. In Chapters 10 through 12, we will discuss our approach to data stream analytics. Our approach has been built upon several of the previous works. Therefore, in this chapter we review the previous works in data stream classification and novelty detection. Also, we discuss related works in semisupervised clustering which is an important component of our data stream classification technique with limited labeled data.