Compression Schemes for Mining Large Datasets: A Machine Learning Perspective

Compression Schemes for Mining Large Datasets: A Machine Learning Perspective

M. Narasimha Murty, T. Ravindra Babu, S. V. Subrahmanya


This publication addresses the demanding situations of information abstraction new release utilizing a least variety of database scans, compressing facts via novel lossy and non-lossy schemes, and engaging in clustering and class without delay within the compressed area. Schemes are offered that are proven to be effective either when it comes to house and time, whereas at the same time supplying a similar or higher type accuracy. positive factors: describes a non-lossy compression scheme in accordance with run-length encoding of styles with binary valued positive factors; proposes a lossy compression scheme that acknowledges a trend as a chain of beneficial properties and choosing subsequences; examines no matter if the identity of prototypes and lines will be completed concurrently via lossy compression and effective clustering; discusses how you can utilize area wisdom in producing abstraction; reports optimum prototype choice utilizing genetic algorithms; indicates attainable methods of facing enormous facts difficulties utilizing multiagent platforms.

Show sample text content

Download sample