Date of Award
Fall 12-2020
Degree Type
Thesis
Degree Name
M.S.
Degree Program
Computer Science
Department
Computer Science
Major Professor
Tamjidul Hoque
Second Advisor
Shaik Arifuzzaman
Third Advisor
Mahdi Abdelguerfi
Fourth Advisor
Warren Wood
Abstract
This work is concerned with the viability of Machine Learning (ML) in training models for predicting global bathymetry, and whether there is a best fit model for predicting that bathymetry. The desired result is an investigation of the ability for ML to be used in future prediction models and to experiment with multiple trained models to determine an optimum selection. Ocean features were aggregated from a set of external studies and placed into two minute spatial grids representing the earth's oceans. A set of regression models, classification models, and a novel classification model were then fit to this data and analyzed. The novel classification model is optimized by selecting the best performing model in a geospatial area. This optimization increases prediction accuracy for test purposes by approximately 3%. These models were trained using bathymetry data from the ETOPO2v2 dataset. Analysis and validation for each model also used bathymetry from the ETOPO dataset, and subsequent metrics were produced and reported. Results demonstrate that ocean features can potentially be used to build a prediction model for bathymetry with the inclusion of accurate data and intelligent model selection. Based on the results in this work, evidence supports that no single model will best predict all Global bathymetry.
Recommended Citation
Moran, Nicholas P., "Machine Learning Model Selection for Predicting Global Bathymetry" (2020). University of New Orleans Theses and Dissertations. 2837.
https://scholarworks.uno.edu/td/2837
Included in
Artificial Intelligence and Robotics Commons, Data Science Commons, Geology Commons, Geophysics and Seismology Commons
Rights
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.