Hybrid Stacked LSTM Based Classification in Prediction of Weather Forecasting Using Deep Learning

Abstract

High dimensionality, interactions on numerous distinct spatial and temporal dimensions, and chaotic dynamics are the dominant factors in weather and climate prediction. This research aims at predicting the weather using classification techniques of deep learning. Here we use Hybrid_BiLSTMtechnique which comprises of both LSTM and Bi-LSTM for classification of data. the data has been pre-processed using standard scaling technique before classification. Here, we offer a method for predicting the weather that uses historical data from numerous weather stations to train basic ML models, which can quickly and accurately forecast specific weather conditions for the near future.