30th Annual Student Research Conference
Efficient and Scalable Categorical Prediction using Neural Networks
Callie E. Warren* and Andrew D. Stratmann
Dr. Robert Matthews, Faculty Mentor
Neural networks are a machine learning algorithm modeled loosely after the human brain. We investigate the use of neural networks in a scaled game of 20 Questions named the Truvoyant project. Truvoyant uses a series of questions to attempt to deduce a term specific to Truman State University imagined by the user. The user interacts with Truvoyant through a web-based interface. The accuracy of Truvoyant's guesses is recorded before and after exposure to training data to analyze the effect of backpropagation, an accepted learning mechanism. In addition, various sizes of training data sets are used to analyze the optimal extent of training.
Keywords: Neural Network, Artificial Intelligence, Machine Learning, Computer Science, Mathematics, Truman State University
Topic(s):Computer Science
Mathematics
Presentation Type: Special Request
Session: 15-
Location: GEO - SUB
Time: