Rainfall prediction using Back Propagation Neural Network Model with Improved Flower Pollination Optimization Algorithm

Abstract

Rainfall prediction is the recent research as it set up the farmers to move with the effectual decision-making regarding agriculture both in irrigation and cultivation. The conventional prediction techniques are daunting, the rainfall prediction depends upon three main factors such as rainfall, humidity, and rainfall recorded in the preceding years that ensued in enormous time-consumption and leverages enormous computational efforts related with the evaluation. Hence, this work adopts the rainfall prediction model based on the deep learning network: Back Propagation Neural Network system. The weights of deep learning are tuned optimally by exploiting the Improved Flower Pollination Algorithm to ease the global optimal tuning of the weights and promise improved prediction accuracy. Conversely, the developed deep learning model is modeled in the MapReduce model which set up the effectual handling of the big data.

References