Comparison of Enhanced XGBoost Algorithm with Light Gradient Boosting Machine to Determine the Prediction of Black Friday Sales

Koyyala Ramprasad, R. Rajasekaran

2023

Abstract

This research evaluates the efficacy of the Enhanced XGBoost algorithm in forecasting sales during the Black Friday event, juxtaposed against the Light Gradient Boosting Machine’s performance. With a focus on enhancing sales prediction precision, the study uses both the Enhanced XGBoost and the Light Gradient Boosting Machine. Employing a sample size of 10 for each, determined using ClinCalc software with a confidence interval of 95% and an alpha value of 0.05, the investigation relies on a sales analysis dataset from Kaggle, comprising 80,550 entries. SPSS statistical analysis reveals the Enhanced XGBoost’s accuracy stands at 93%, outstripping the Light Gradient Boosting Machine’s 73%. Furthermore, a significant difference is observed between the two, with a p-value of 0.001. The findings clearly show the Enhanced XGBoost’s superior performance in Black Friday sales predictions.

Paper Citation

in Harvard Style

Ramprasad K. and Rajasekaran R. (2023). Comparison of Enhanced XGBoost Algorithm with Light Gradient Boosting Machine to Determine the Prediction of Black Friday Sales. In Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT; ISBN 978-989-758-661-3, SciTePress, pages 474-479. DOI: 10.5220/0012518100003739

in Bibtex Style

@conference{ai4iot23,
author={Koyyala Ramprasad and R. Rajasekaran},
title={Comparison of Enhanced XGBoost Algorithm with Light Gradient Boosting Machine to Determine the Prediction of Black Friday Sales},
booktitle={Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT},
year={2023},
pages={474-479},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012518100003739},
isbn={978-989-758-661-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT
TI - Comparison of Enhanced XGBoost Algorithm with Light Gradient Boosting Machine to Determine the Prediction of Black Friday Sales
SN - 978-989-758-661-3
AU - Ramprasad K.
AU - Rajasekaran R.
PY - 2023
SP - 474
EP - 479
DO - 10.5220/0012518100003739
PB - SciTePress