36th Annual Student Research Conference
The Relevance of Accurate Data in the Process of Training an Artificial Neural Network
Tyler J. Bryan
Dr. Ruthie Halma, Faculty Mentor
An artificial neural network (ANN) was trained to predict the price of a car based on 6 features from a database of over 10,000 used cars. The process of coding an ANN using Python will be discussed. Multiple tests of various scenarios were run including missing or incorrect data during the ANN implementation process. Accuracy of the output was then compared to the corresponding results when inputting the complete and correct data. When comparing the best and worst scenario, the latter had a 572% higher error loss than the best, which had a 0.14% error loss.
Keywords: Machine Learning, Artificial Neural Network
Topic(s):Computer Science
Presentation Type: Oral Presentation
Session: 209-4
Location: MG 1098
Time: 11:00