TY - JOUR
T1 - The Benefits of Automated Machine Learning in Hospitality
T2 - A Step-By-Step Guide and AutoML Tool
AU - Castelli, Mauro
AU - Pinto, Diego Costa
AU - Shuqair, Saleh
AU - Montali, Davide
AU - Vanneschi, Leonardo
N1 - info:eu-repo/grantAgreement/FCT/3599-PPCDT/DSAIPA%2FDS%2F0113%2F2019/PT#
info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F04152%2F2020/PT#
Castelli, M., Pinto, D. C., Shuqair, S., Montali, D., & Vanneschi, L. (2022). The Benefits of Automated Machine Learning in Hospitality: A Step-By-Step Guide and AutoML Tool. Emerging Science Journal, 6(6), 1237-1254. https://doi.org/10.28991ESJ-2022-06-06-02. Funding:This study was supported by grant DSAIPA/DS/0113/2019 from FCT (Fundação para a Ciência e a Tecnologia), Portugal. This work was also supported by national funds through FCT (Fundação para a Ciência e a Tecnologia), under the project - UIDB/04152/2020 - Centro de Investigação em Gestão de Informação (MagIC)/NOVA IMS.
PY - 2022/12/1
Y1 - 2022/12/1
N2 - The manuscript presents a tool to estimate and predict data accuracy in hospitality by means of automated machine learning (AutoML). It uses a tree-based pipeline optimization tool (TPOT) as a methodological framework. The TPOT is an AutoML framework based on genetic programming, and it is particularly useful to generate classification models, for regression analysis, and to determine the most accurate algorithms and hyperparameters in hospitality. To demonstrate the presented tool’s real usefulness, we show that the TPOT findings provide further improvement, using a real-world dataset to convert key hospitality variables (customer satisfaction, loyalty) to revenue, with up to 93% prediction accuracy on unseen data.
AB - The manuscript presents a tool to estimate and predict data accuracy in hospitality by means of automated machine learning (AutoML). It uses a tree-based pipeline optimization tool (TPOT) as a methodological framework. The TPOT is an AutoML framework based on genetic programming, and it is particularly useful to generate classification models, for regression analysis, and to determine the most accurate algorithms and hyperparameters in hospitality. To demonstrate the presented tool’s real usefulness, we show that the TPOT findings provide further improvement, using a real-world dataset to convert key hospitality variables (customer satisfaction, loyalty) to revenue, with up to 93% prediction accuracy on unseen data.
KW - Artificial Intelligence
KW - Automated Machine Learning
KW - Behavioral Research
KW - Hospitality
UR - http://www.scopus.com/inward/record.url?scp=85143207889&partnerID=8YFLogxK
U2 - 10.28991/ESJ-2022-06-06-02
DO - 10.28991/ESJ-2022-06-06-02
M3 - Article
VL - 6
SP - 1237
EP - 1254
JO - Emerging Science Journal
JF - Emerging Science Journal
SN - 2610-9182
IS - 6
ER -
