Data Stream Classification Techniques to Find Novel Class – A Review

Abstract

Data Stream Mining (DSM) is the way of retrieving useful knowledge from fast information records. Today, there are number of application that produce massive amount of stream data. Concept evolution, the phenomenon of class disappearance and emergence is a valuable research topic in data stream mining field. Concept evolution happens when a new class occurs in the stream. Similar to data mining, data stream mining includes classification and clustering techniques; the special focus of this paper is on classification techniques developed to handle data streams. The problem of classification is one of the most widely studied in the context of data stream mining. The problem of classification is made more difficult by the evolution of the underlying data stream. Based on the efficiency of the classification technique used, the accuracy of classification varies. This paper reviews different classification techniques for handling the novel class in the data streams.

Authors and Affiliations

Aswathi R Kaimal, Joshy Thomas Jose

Keywords

Related Articles

Optimization of machining parameters during the steel work-pieces turning

One of the mostly used processes in the manufacturing machine parts, is turning process. The parameters of this process have different effects on the machining capabilities. So one of the problems associated with this p...

Unpropitious Effect of Bisphenol-A and Its Impact on Human Health

Bisphenol A (BPA) is a chemical used in the lining of some food and beverage packaging to protect food from contamination and extend shelf life. It’s also used in non-food products. It has been found that BPA has the po...

Plant Growth Hormones - The Key Players In The Process Of Seed Germination

Plant growth hormones along with numeral factors are the key players in the process of seed germination and seedling growth. They are produced by both plants and soil bacteria. They are defined as organic compounds othe...

Modelling of MIB Based SPLDs for Future Computing

Designing low power ultra dense digital circuits are considered both fascinating and challenging technology in post CMOS era. SET (Single Electron Transistor) technology came up as a promising candidate and is largely a...

A Scalable Approach To Detect A Duplicate Data Using PSNM And PB Algorithm

Now a day if we consider a data set we can find more duplicate data. Determining the redundant data in the data server is an open research in the data intensive application. The traditional method detects the duplicate...

Download PDF file
  • EP ID EP24523
  • DOI -
  • Views 306
  • Downloads 9

How To Cite

Aswathi R Kaimal, Joshy Thomas Jose (2017). Data Stream Classification Techniques to Find Novel Class – A Review. International Journal for Research in Applied Science and Engineering Technology (IJRASET), 5(6), -. https://europub.co.uk/articles/-A-24523