Meta-BCS: a novel way to BCS ranking using generalized regression neural network and genetic algorithm

  • Authors:
  • Lei Zhang;Haiquan Chen

  • Affiliations:
  • Auburn University, AL;Auburn University, AL

  • Venue:
  • Proceedings of the 46th Annual Southeast Regional Conference on XX
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we attempt to run vulnerability analysis on the Bowl Championship Series (BCS) ranking system. We used Generalized Regression Neural Network (GRNN) to create a model for each poll and then applied these models to simulate a random season. After this, a modified genetic algorithm (GA) is applied to evolve schedules that minimize the difference between the number 2 ranked team and the number 3 ranked team. By simulating the seasons, we aim to show that current BCS system in fact converges faster than the proposed meta-BCS system.