Turning the blackbox into a glassbox: An explainable machine learning approach for understanding hospitality customer
Travel and hospitality industry are adopting high end technology to reach out their customers. New levels of disruption are being introduced using Artificial Intelligence and Machine Learning techniques of research. In the realm of physical and tangible aspects of service, since this sector have vas...
Guardado en:
Autores principales: | , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
Elsevier
2021
|
Materias: | |
Acceso en línea: | https://doaj.org/article/5f940ef69e7d4012b67be0e2d867977e |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:5f940ef69e7d4012b67be0e2d867977e |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:5f940ef69e7d4012b67be0e2d867977e2021-12-04T04:36:42ZTurning the blackbox into a glassbox: An explainable machine learning approach for understanding hospitality customer2667-096810.1016/j.jjimei.2021.100050https://doaj.org/article/5f940ef69e7d4012b67be0e2d867977e2021-11-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2667096821000434https://doaj.org/toc/2667-0968Travel and hospitality industry are adopting high end technology to reach out their customers. New levels of disruption are being introduced using Artificial Intelligence and Machine Learning techniques of research. In the realm of physical and tangible aspects of service, since this sector have vast amounts of data available application of these techniques are being tested to explore various aspect of the customer experience. Data analysis algorithms like machine learning have been used by firms across the world to analyze these data and understand various facets of customer experience. However, these algorithms suffer from blackbox issues where businesses do not know how the outcome was achieved and if the analysis is biased in some way. In this study, we utilize an explainable machine learning approach on Airbnb data to understand solve two challenges associated with large-scale marketing data. First, we identify and prioritize the customer experience factors like bedrooms, host status, host response rate etc. that have an impact on product pricing. Secondly, we build and test an explainable predictive model on the same. Our research has significant implications for academia and industry in identifying applications of explainable algorithms in decision-making.Ritu SharmaArpit KumarCindy ChuahElsevierarticleCustomer experienceArtificial intelligenceMachine learningExplainable AITourismInformation technologyT58.5-58.64ENInternational Journal of Information Management Data Insights, Vol 1, Iss 2, Pp 100050- (2021) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
Customer experience Artificial intelligence Machine learning Explainable AI Tourism Information technology T58.5-58.64 |
spellingShingle |
Customer experience Artificial intelligence Machine learning Explainable AI Tourism Information technology T58.5-58.64 Ritu Sharma Arpit Kumar Cindy Chuah Turning the blackbox into a glassbox: An explainable machine learning approach for understanding hospitality customer |
description |
Travel and hospitality industry are adopting high end technology to reach out their customers. New levels of disruption are being introduced using Artificial Intelligence and Machine Learning techniques of research. In the realm of physical and tangible aspects of service, since this sector have vast amounts of data available application of these techniques are being tested to explore various aspect of the customer experience. Data analysis algorithms like machine learning have been used by firms across the world to analyze these data and understand various facets of customer experience. However, these algorithms suffer from blackbox issues where businesses do not know how the outcome was achieved and if the analysis is biased in some way. In this study, we utilize an explainable machine learning approach on Airbnb data to understand solve two challenges associated with large-scale marketing data. First, we identify and prioritize the customer experience factors like bedrooms, host status, host response rate etc. that have an impact on product pricing. Secondly, we build and test an explainable predictive model on the same. Our research has significant implications for academia and industry in identifying applications of explainable algorithms in decision-making. |
format |
article |
author |
Ritu Sharma Arpit Kumar Cindy Chuah |
author_facet |
Ritu Sharma Arpit Kumar Cindy Chuah |
author_sort |
Ritu Sharma |
title |
Turning the blackbox into a glassbox: An explainable machine learning approach for understanding hospitality customer |
title_short |
Turning the blackbox into a glassbox: An explainable machine learning approach for understanding hospitality customer |
title_full |
Turning the blackbox into a glassbox: An explainable machine learning approach for understanding hospitality customer |
title_fullStr |
Turning the blackbox into a glassbox: An explainable machine learning approach for understanding hospitality customer |
title_full_unstemmed |
Turning the blackbox into a glassbox: An explainable machine learning approach for understanding hospitality customer |
title_sort |
turning the blackbox into a glassbox: an explainable machine learning approach for understanding hospitality customer |
publisher |
Elsevier |
publishDate |
2021 |
url |
https://doaj.org/article/5f940ef69e7d4012b67be0e2d867977e |
work_keys_str_mv |
AT ritusharma turningtheblackboxintoaglassboxanexplainablemachinelearningapproachforunderstandinghospitalitycustomer AT arpitkumar turningtheblackboxintoaglassboxanexplainablemachinelearningapproachforunderstandinghospitalitycustomer AT cindychuah turningtheblackboxintoaglassboxanexplainablemachinelearningapproachforunderstandinghospitalitycustomer |
_version_ |
1718372913507205120 |