Evaluating Data Minability Through Compression - An Experimental Study

The authors' goal is to show that compression can be used as a tool to evaluate the potential of a data set of producing interesting results in a data mining process. The basic idea that data that displays repetitive patterns or patterns that occur with a certain regularity will be compressed more efficiently compared to data that has no such characteristics. Thus, a pre-processing phase of the mining process should allow to decide whether a data set is worth mining, or compare the interestingness of applying mining algorithms to several data sets.

Provided by: IARIA Topic: Data Management Date Added: Sep 2012 Format: PDF

Find By Topic