Date of Award

Summer 8-2020

Degree Type


Degree Name


Degree Program

Computer Science


Computer Science

Major Professor

Hoque, Tamjidul; Abdelguerfi, Mahdi

Second Advisor

Ioup, Elias

Third Advisor

Arifuzzaman, Shaikh


This research solely focuses on understanding and predicting weather behavior, which is one of the important factors that affect airplanes in flight. The future weather information is used for informing pilots about changing flight conditions. In this paper, we present a new approach towards forecasting one component of weather information, wind speed, from data captured by airplanes in flight. We compare NASA’s ACT-America project against NOAA’s Wind Aloft program for prediction suitability. A collinearity analysis between these datasets reveals better model performance and smaller test error with NASA’s dataset. We then apply machine learning and a genetic algorithm to process the data further and arrive at a competitive error rate. The sliding window approach is used to find the best window size, and then we create a forecasting model that predicts wind speed at high altitudes 10 mins ahead of time. Finally, a stacking-based framework was used for better performance than individual learning algorithms to get root means square error (RMSE) of the best combination as 0.674, which is 98.4% better than the state-of-the-art approach.


The University of New Orleans and its agents retain the non-exclusive license to archive and make accessible this dissertation or thesis in whole or in part in all forms of media, now or hereafter known. The author retains all other ownership rights to the copyright of the thesis or dissertation.