Rough Sets base Incremental Associative Classification Rules Generation on MapReduce Framework

  • Hanumanthu Bhukya, Dr. M.Sadanandam

Abstract

MapReduce is a well-proven computing paradigm for Bigdata analytics, which is the latest buzz word in the data science community. Big data challenges the world of computing not only by velocity, but also by veracity and velocity. Veracity refers the handling deviated information where the velocity the regular improvements brought to big data. In the existing literature, it can be observed that research only pays attention to the handling of Bigdata's volume properties, avoiding other important properties of veracity and velocity. In order to overcome these problems, our model proposes a rough set classification model based on the MapReduce framework for the handling of veracity properties. At the same time, the proposed rough set based classification model for the MapReduce framework is also extended to handle data increments to counter Bigdata's data velocity properties. Two algorithms was proposed where the first one guards the veracity of big data then the second algorithm handles velocity property by handling incremented data. The proposed work was tested on Bigdata benchmark data sets, which demonstrated the importance of the proposed system for handling incremental data from big data systems using a rough set based classification model for the MapReduce framework.

Published
2020-06-06
How to Cite
Hanumanthu Bhukya, Dr. M.Sadanandam. (2020). Rough Sets base Incremental Associative Classification Rules Generation on MapReduce Framework. International Journal of Advanced Science and Technology, 29(05), 13218-13227. Retrieved from http://sersc.org/journals/index.php/IJAST/article/view/25949