Is this project an undergraduate, graduate, or faculty project?
Graduate
group
What campus are you from?
Daytona Beach
Authors' Class Standing
Tyler Wise, Graduate Student Keely Mashburn, Senior Alani Seaman, Junior
Lead Presenter's Name
Tyler Wise
Faculty Mentor Name
Khem Poudel
Abstract
By applying the concepts of machine learning, the aim is to create a program that utilizes neural networks to analyze the wait times at various Florida theme parks. These parks include SeaWorld, Busch Gardens, both Universal parks, and all four Walt Disney World parks. The project hinges on a distributed computing architecture that divides the work as assigned by a master, rather than strictly parallelizing the code. The technology used throughout this project is hosted on Amazon Web Services, utilizing their Relational Database Service and Sagemaker platforms. MySQL, Python, and Tensorflow are the core software technologies running on this infrastructure. Each of these programs plays a role in creating a complete solution towards creating a recurrent neural network that delivers a list of wait times synthesized for the following hour that users can benefit from in real time. For our analysis of the network's validity, we will create a statistical distribution for the error present in each ride’s prediction. This will be performed on a testing data set, which is composed of twenty percent of the overall data chosen at random.
Did this research project receive funding support from the Office of Undergraduate Research.
No
Using Recurrent Neural Networks in a Distributed Computing Environment for Predicting Time-Variant Data
By applying the concepts of machine learning, the aim is to create a program that utilizes neural networks to analyze the wait times at various Florida theme parks. These parks include SeaWorld, Busch Gardens, both Universal parks, and all four Walt Disney World parks. The project hinges on a distributed computing architecture that divides the work as assigned by a master, rather than strictly parallelizing the code. The technology used throughout this project is hosted on Amazon Web Services, utilizing their Relational Database Service and Sagemaker platforms. MySQL, Python, and Tensorflow are the core software technologies running on this infrastructure. Each of these programs plays a role in creating a complete solution towards creating a recurrent neural network that delivers a list of wait times synthesized for the following hour that users can benefit from in real time. For our analysis of the network's validity, we will create a statistical distribution for the error present in each ride’s prediction. This will be performed on a testing data set, which is composed of twenty percent of the overall data chosen at random.