Key points are not available for this paper at this time.
Data mining is one of the most exciting fields of research for the researcher. As data is getting digitized, systems are getting connected and integrated, scope of data generation and analytics has increased exponentially. Today, most of the systems generate non-stationary data of huge, size, volume, occurrence speed, fast changing etc. these kinds of data are called data streams. One of the most recent trend i.e. IOT (Internet Of Things) is also promising lots of expectation of people which will ease the use of day to day activities and it could also connect systems and people together. This situation will also lead to generation of data streams, thus present and future scope of data stream mining is highly promising. Characteristics of data stream possess many challenges for the researcher; this makes analytics of such data difficult and also acts as source of inspiration for researcher. Outlier detection plays important role in any application. In this paper we reviewed different techniques of outlier detection for stream data and their issues in detail and presented results of the same.
Shukla et al. (Tue,) studied this question.