Real-Time Bigdata Analytics: A Stream Data Mining Approach

Abstract

The outburst of Bigdata has driven a great deal of research to build and extend systems for in-memory data analytics in real-time environment. Stream data mining makes allocation of tasks efficient among various distributed computational resources. Managing chunk of unbounded stream data is challenging task as data ranges from structured to unstructured. Beyond size, it is heterogeneous and dynamic in nature. Scalability and low-latency outputs are vital while dealing with big stream data. The potentiality of traditional approach like data stream management systems (DSMSs) is inadequate to ingest and process huge volume and unbounded stream data for knowledge extraction. A novel approach develops architectures, algorithms, and tools for uninterrupted querying over big stream data in real-time environment. This paper overviews various challenges and approaches related to big stream data mining. In addition, this paper surveys various platforms and proposed framework which can be applied to near-real- or real-time applications.