Leena A Deshpande*

Department of Computer, Vishwakarma Institute of Information Technology, Pune, India

*Corresponding Author: Leena A Deshpande, Department of Computer, Vishwakarma Institute of Information Technology, Pune, India.

Received: December 01, 2022; Published: January 10, 2023

Abstract

In today’s era, increase in volume of data and due to variety of patterns generated, multiple challenges are raised and they need to be tackled using models based on classification theory. Many current applications face these challenges which need to be tackled with advanced solutions. Recent applications like sensor data, network traffic monitoring, stock market predictions, call centre records, web log analysers, and chemical reactor plants etc. process high amount of data where arrived data distribution may get change after certain period of time. In social media applications, users comment and share their views on social media like Twitter, Facebook causes drastic change in the behaviour and pattern as user may change their perspective and change their opinion or suddenly more hit for a particular topic or post may arrive. Such a data is referred as Data Stream. Data Stream: To handle such concept drifts in the arriving data, data stream classification is applied as a novel research problem which leads to identify change in arriving pattern Popular algorithms of data stream mining are Classification, Clustering and Frequent pattern mining Such novel problems opens the research challenges which addresses rigorous training of streaming data, optimum selection of algorithms, new feature selection which must be incorporated in existing machine learning algorithms.

Keywords: Data Stream; Clustering; Stock Market

References

1. J Gama. “A survey on concept drift adaptation”. ACM Computing Surveys4 (2014): 1-37.

Citation

Citation: Leena A Deshpande. “Concept Drift in Machine Learning". Acta Scientific Computer Sciences 5.2 (2023): 25.

Copyright

Copyright: © 2023 Leena A Deshpande. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.