A Distributed Framework for Predictive Analytics Using Big Data and MapReduce Parallel Programming

Algorithm 1

Map function of MapReduce-I

	Function MAP-1 (training dataset)
	Begin
	Input: training dataset D with m instances and n attributes
	Partition the dataset D into s partitions as p1, p2, p3….. ps
	Read x_train[], y_train for each partitioned dataset
	Compute intercept and correlation coefficients for each block of instances
	Convert it into (key, value > pair as < Dataset_id, (intercept, coefficients)>
	Output < Dataset_id, <(intercept, coefficients[])>
	end