Abstract by Stephanie Herron

Personal Infomation

Presenter's Name

Stephanie Herron

Degree Level


Abstract Infomation


Physics and Astronomy

Faculty Advisor

Tracianne Neilsen


Support Vectors Machine for Underwater Acoustic Signal Classification


While machine learning has become increasingly popular as a means to learn information from large datasets, the question remains how different machine learning models can best be used to improve SONAR operations. In the current research, machine learning is used to predict the distance and depth of a source and the seabed type given a set of acoustic signals in an underwater ocean environment. For this classification problem, a model is being developed using a Support Vector Machine. This model finds natural divisions in the data to divide it into multiple different classes using hyperplanes, which maximize the division between data samples at the extremes of classes. By using these divisions, the model can then predict the classes for testing data. Future work will continue to compare the efficiency and accuracy of Support Vector Machines against other types of Machine Learning in underwater acoustics.