Tweets provide real-time insights and sentiment from a diverse range of sources, offering data for understanding public opinion, emergencies, and trends. This research focuses on the applications of the Natural Language Processing (NLP) techniques for disaster related tweets. Data preprocessing is performed with the help of NLP techniques such as Bag-of-words (BoW) and term frequency-inverse document frequency (TF-IDF) approaches are used for feature extraction. This paper highlights the elements for creating a model using traditional algorithms, including Logistic Regression, Random Forest Classifier and Support Vector Classifier, were used to classify tweets into disaster and non-disaster categories which resulted in a accuracy of 82%. To further improve the accuracy of our model, we implemented modern algorithms, such as Extreme Gradient Boosting (Xgboost), a Recurrent Neural Network (RNN) model with Long Short-term Memory (LSTM) and a LSTM model was implemented, demonstrated an accuracy of 96.32%.