Is this project an undergraduate, graduate, or faculty project?
Undergraduate
Project Type
group
Campus
Daytona Beach
Authors' Class Standing
Emily Diegel, Sophmore Yianni Paraschos,Sophmore Annaelise Swanson, Sophmore Taryn Trimble, Junior
Lead Presenter's Name
Emily Diegel
Lead Presenter's College
DB College of Arts and Sciences
Faculty Mentor Name
Mihhail Berezovski
Abstract
Machine learning has become a common tool within the tech industry due to its high versatility and efficiency with large datasets. Partnering with the Nevada National Security Site, our goal is to improve accuracy of machine predictions by utilizing deep learning, which will enable the power and accuracy of a prediction to grow from the model. To build a deep learning model, multiple neural network architectures were developed and combined to create an ensemble neural network. The project’s objective is to determine the comparative differences between the efficiency of the ensemble neural network versus each individual neural network. The data set used to test, validate, and train the networks is 1D regressive. After testing architecture and determining accuracy of certain networks, the model will be updated and tested again to compare accuracies. Accuracy is the number of correct predictions over the total number of predictions. As model precision is a key aspect of machine learning, emphasis is placed on the efficiency of ensemble neural networks.
Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?
No
Ensemble Deep Learning
Machine learning has become a common tool within the tech industry due to its high versatility and efficiency with large datasets. Partnering with the Nevada National Security Site, our goal is to improve accuracy of machine predictions by utilizing deep learning, which will enable the power and accuracy of a prediction to grow from the model. To build a deep learning model, multiple neural network architectures were developed and combined to create an ensemble neural network. The project’s objective is to determine the comparative differences between the efficiency of the ensemble neural network versus each individual neural network. The data set used to test, validate, and train the networks is 1D regressive. After testing architecture and determining accuracy of certain networks, the model will be updated and tested again to compare accuracies. Accuracy is the number of correct predictions over the total number of predictions. As model precision is a key aspect of machine learning, emphasis is placed on the efficiency of ensemble neural networks.