High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Modified Map Reduce Model using Fuzzy Neural Network


Vatsal Gor , Nsit Jetalpur


BigData, fuzzy Neural Network, ANFIS


The MapReduce paradigm is now the de facto standard for processing and generation large scale datasets. Generally in MapReduce paradigm, user specifies a map function that processes a key/value pair to generate a set of intermediate key-value pairs & reduce function which merges all intermediate values associated with the same intermediate key. This paradigm is enough to solve many real world problems. Here fuzzy neural network is applied in the MapReduce paradigm to improve its efficiency. It is also referred as Neuro-fuzzy which is combination of artificial neural network and fuzzy logic. Now, here the key/value generated pair via map function is feed as input in neuro fuzzy system or model to improve efficiency of current map reduce paradigm. The proposed Fuzzy Neural Network model aims at training the fuzzy Neural Network model in MapReduce programming model. Since the MapReduce programming model has the ability to rapidly process large amount of data in parallel and can use the outcome of training process for the prediction purpose. This can produce better prediction accuracy. In order to train large data in distributed mode it can be effective in reducing the processing time.

Other Details

Paper ID: IJSRDV3I1365
Published in: Volume : 3, Issue : 1
Publication Date: 01/04/2015
Page(s): 206-207

Article Preview

Download Article