This repository contains the source code of paper: "Semantic-based End-to-End Learning for Typhoon Intensity Prediction"

Our project aims to help people come up with solutions to cope up with disasters before, during and after the disaster. We incorporate internet-less chatting, skin detection, fundraising, disaster prediction, live ground status and drones to deliver amenities into our project.

A Bidirectional LSTM model to classify whether a given tweet talks about a real disaster or not. This was my project in "CSC 522: Automated Learning and Data Analysis" course at NC State University.

An approach to solve the Kaggle Competition, Natural Language Processing with Disaster Tweets

A disaster prediction web application aided with AI to provide user with information on any upcoming natural disasters noted

A comprehensive, scientific-grade Python application for earthquake risk assessment that integrates geological, tectonic, and volcanic analysis using advanced machine learning techniques.

Computer vision system that analyzes geological formations from satellite/drone imagery to identify mineral deposits and predict natural disasters.

An advanced AI-powered disaster prediction platform utilizing a weighted fusion of LSTM, Prophet, and Random Forest models. Implements a high-accuracy ensemble mechanism for global hazard forecasting as presented in Research Paper : "AI-Powered Disaster Prediction System: A Comprehen sive Approach to Mitigating Natural Disasters"

A website that users can use to detect if their uploaded image is disasterous or not using CNN.

Deep learning system for climate change analysis using satellite imagery and weather data. Predicts natural disasters, monitors deforestation, tracks glacier melting, and analyzes urban heat islands.

This project involves building a machine learning model to classify tweets as disaster-related or not, using a dataset of 10,000 hand-labeled tweets to assist in real-time emergency detection.