Chi-FRBCS-BigData-Max: MapReduce implementation of the Chi et al.’s approach.
Switch branches/tags
Nothing to show
Clone or download
Latest commit 4b9a7f5 Jun 19, 2015

#The Chi-FRBCS-BigData algorithm: A MapReduce Design based on the Fusion of Fuzzy Rules

The Chi-FRBCSBigData algorithm uses two different MapReduce processes. One MapReduce process is devoted to the building of the model from a big data training set (Figure 1). The other MapReduce process is used to estimate the class of the examples belonging to big data sample sets using the previous learned model (Figure 2). Both parts follow the MapReduce design, distributing all the computations along several processing units that manage different chunks of information, aggregating the results obtained in an appropriate manner. We have developed two versions named Chi-FRBCS-BigData-Max and Chi-FRBCSBigData-Ave which share most of their operations, however, they behave differently in the reduce step of the approach, when the different RBs generated by each map are fused. These versions obtain different RBs and thus, different KBs.

Figure 1: A flowchart of how the building of the KB is organized in Chi-FRBCS-BigData

Figure 2: flowchart of how the classification of a big dataset is organized in Chi-FRBCS-BigData


S. Río, V. López, J.M. Benítez, F. Herrera. A MapReduce Approach to Address Big Data Classification Problems Based on the Fusion of Linguistic Fuzzy Rules. International Journal of Computational Intelligence Systems 8:3 (2015) 422-437. doi: 10.1080/18756891.2015.1017377 link to pdf file