MFTCN: A Multi-Frequency Temporal Convolutional Network for Short-Term Temperature Prediction

Temperature and Dewpoint Forecasting using the ARIMA and Auto-ARIMA model respectively.

In this project, I analyzed weather data from the NCEI Global Surface Summary of Day dataset using PySpark in Jupyter Notebook. Tasks included data cleaning, statistical analysis, and forecasting for temperature, wind speed, precipitation, and extreme weather events. The project also predicts future weather patterns for Cincinnati and Florida.

World Temperature Viewer is more than just a data visualization tool; it is a conduit for knowledge transfer, fostering a greater understanding of the far-reaching impacts of global warming and climate change

I have worked on this project as part of my exploration of machine learning algorithms. I have understood the deep concept behind Arima-Model and Auto-Arima-Model. Also, this notebook contains data cleaning and data visualization things which make things easier to understand.

Physics-informed ensemble for 12-h city-region temperature forecasts. Advection–diffusion prior + ConvLSTM + RAFL + edRVFL-SC for extreme-event warnings.

World temperature analysis and forecasting with ARIMA variants and Prophet.

"Correcting bias in numerical weather forecasts using regression models and real-world environmental data."

A project that applies machine learning, deep learning, and forecasting techniques to predict daily U.S. temperatures, enhanced with feature engineering, visualizations, and interactive dashboards to support climate and environmental decision-making.

This repository contains the code and resources for the Data Powered Software Solutions course (IT3212) at the Norwegian University of Science and Technology (NTNU) during Fall 2024. The course focuses on leveraging data to build efficient and effective software solutions. The projects have been developed as a group effort.