Feature Selection Using a Metaheuristic Method

• Activity Number: 210 - SLDS CSpeed 3
• Type: Contributed
• Date/Time: Tuesday, August 10, 2021 : 1:30 PM to 3:20 PM
• Sponsor: Korean International Statistical Society
• Author(s): Myung Soon Song* and Francis J Vasko and Yun Lu and Kyle Callaghan
• Companies: Kuztown University of Pennsylvania and Kuztown University of Pennsylvania and Kuztown University of Pennsylvania and Kuztown University of Pennsylvania
• Keywords: feature selection; regression; metaheuristic; optimization; jaya

Abstract

Feature or variable selection is one of the very popular topics from regression models. Besides many conventional approaches, some metaheuristic approaches from the realm of optimization including, but are not limited to, GA (Genetic Algorithm) or SA (simulated annealing) have been suggested so far. These methods have a considerable advantage over the conventional methods when dealing with a variety of problems, but how to fine-tune parameters is very challenging. In this research, a parameter-free approach called Jaya will be suggested and explored. Many methods such as GA, TBO (Teaching Based Optimization), and Jaya will be compared to one another with a real-world dataset and a simulated dataset. The impact of using local search will also be analyzed. This is a follow-up to what we presented at JSM 2020.