Research and Parameter Setting Recommendations for the UCB Algorithm Based on Advertisement Deployment on the Amazon Website

Yuankun Zhou

2024

Abstract

The multi-armed bandit problem, a cornerstone of decision-making theory, primarily addresses the critical balance between exploration and exploitation to optimize choices. This problem is intrinsically linked with numerous real-world applications, spanning diverse sectors from business to healthcare. It has inspired a variety of algorithms, among which the Upper Confidence Bound (UCB) algorithm stands out. The UCB algorithm, noted for its approach of setting an upper confidence interval for each option as a decision criterion, has captured the attention of many scholars. This paper contextualizes its discussion within the framework of advertisement deployment for products on Amazon, employing various algorithms to simulate ad campaigns and analyze their effectiveness. It summarizes the unique characteristics and appropriate contexts for each algorithm and explores enhancements through structural and parameter adjustments. Based on extensive experimental data, the study offers recommendations for algorithm parameter settings in various scenarios, aimed at maximizing the practical application and effectiveness of these algorithms in real-world settings. This research not only enhances understanding of algorithmic adaptations but also provides valuable insights for their application across different operational environments.

Paper Citation

in Harvard Style

Zhou Y. (2024). Research and Parameter Setting Recommendations for the UCB Algorithm Based on Advertisement Deployment on the Amazon Website. In Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI; ISBN 978-989-758-713-9, SciTePress, pages 685-691. DOI: 10.5220/0012968400004508

in Bibtex Style

@conference{emiti24,
author={Yuankun Zhou},
title={Research and Parameter Setting Recommendations for the UCB Algorithm Based on Advertisement Deployment on the Amazon Website},
booktitle={Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI},
year={2024},
pages={685-691},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012968400004508},
isbn={978-989-758-713-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI
TI - Research and Parameter Setting Recommendations for the UCB Algorithm Based on Advertisement Deployment on the Amazon Website
SN - 978-989-758-713-9
AU - Zhou Y.
PY - 2024
SP - 685
EP - 691
DO - 10.5220/0012968400004508
PB - SciTePress