Big Data Preprocessing by Julián Luengo & Diego García-Gil & Sergio Ramírez-Gallego & Salvador García & Francisco Herrera

Author:Julián Luengo & Diego García-Gil & Sergio Ramírez-Gallego & Salvador García & Francisco Herrera
Language: eng
Format: epub
ISBN: 9783030391058
Publisher: Springer International Publishing

5.3 MRPR: A MapReduce Solution for Prototype Reduction in Big Data Classification

In this section, we will describe the MRPR framework (MapReduce for prototype reduction), a new distributed framework for PR based on the stratification procedure [43]. This framework was designed to tackle the drawbacks associated with stratification: high memory consumption and high complexity, and a poor joining process. MRPR relies on MapReduce to parallelize the PR process and the subsequent fusion process. Concretely, the map phase contains the splitting procedure and the local application of PR. The reduce stage performs a filtering or fusion of prototypes in order to prevent the inclusion of negative prototypes in the final set. Figure 5.1 depicts a simplified scheme of the MRPR framework.

Fig. 5.1MRPR processing scheme. The rectangles represent the reduction and joining processes, and the circles the partial and final reduced sets

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.