Implemented a model that aims to predict whether the input tweet is about a real disaster or not Performed EDA, and pre-processing techniques to clean the raw data. Used GloVe Embedding to convert textual data into numerical format. Used Classification algorithms and LSTM based Neural Network for binary classification Employed BERT, for generating sentence embedding of input, modelled Multi-layered Dense Neural Network for final classification. Achieved validation accuracy of 81.8% and public F1 Score of 82.80%