Exploring the Silver Screen: A Comparative Study of UCB and EUCBV Algorithms in Movie Genre Recommendation

Yixian Chen

2024

Abstract

The study explored the comparison of the efficiency between the UCB algorithm and its enhanced version EUCBV algorithm in the framework of the multi-arm bandit problem. The study first introduced the basic concept of the two algorithms and explained how the algorithm balances the exploration and the exploitation when solving the MAB problem. Then I use detailed mathematical formulas to prove the basic concept of each algorithm while explaining how some important arguments were calculated and their effect on the process of the algorithm. The study not only discusses the theories of the algorithms, but also involves the test of the performance and the efficiency of the algorithms that I apply UCB and EUCBV in real-world situations. The test is about using the algorithm to explore the affection of the movie’s genres to the audience’s rating, and how to use the algorithm’s result to design a recommendation system to gain higher ratings. I compare the results and the performance of the two algorithms, which show that by adding a variance term, the EUCBV’s final reward is much more accurate than the traditional UCB algorithm. Overall, the study fully explored the algorithms’ concept and the comparison of the performance when applied the algorithm in real-world situations, which provides me hard evidence of which algorithm is more efficient in some particular cases.

Download


Paper Citation


in Harvard Style

Chen Y. (2024). Exploring the Silver Screen: A Comparative Study of UCB and EUCBV Algorithms in Movie Genre Recommendation. 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 257-263. DOI: 10.5220/0012925600004508


in Bibtex Style

@conference{emiti24,
author={Yixian Chen},
title={Exploring the Silver Screen: A Comparative Study of UCB and EUCBV Algorithms in Movie Genre Recommendation},
booktitle={Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI},
year={2024},
pages={257-263},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012925600004508},
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 - Exploring the Silver Screen: A Comparative Study of UCB and EUCBV Algorithms in Movie Genre Recommendation
SN - 978-989-758-713-9
AU - Chen Y.
PY - 2024
SP - 257
EP - 263
DO - 10.5220/0012925600004508
PB - SciTePress