Prediction of Hotel Booking Cancellation using CRISP-DM

Andriawan, Zharfan Akbar; Purnama, Satriawan Rasyid; Darmawan, Adam Sukma; Ricko,; Wibowo, Adi; Sugiharto, Aris; Wijayanto, Feri

doi:10.1109/icicos51170.2020.9299011

Cited by 7 publications

(5 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Saputro and Nanang (2021) conducted HOCP using random forest (RF) and extra tree classifiers and identified 79% of all hospitality order cancellers. Andriawan et al (2020) found that RF is more accurate in HOCP than XGBoost, CatBoost and light gradient boosting machine (LightGBM) in HOCP. Satu et al (2020) compared the classification accuracy of XGBoost, gradient boosting, RF, decision tree, LR, KNN and Gaussian Naive Bayes classifier (GNBC) in terms of statistical indicators.…”

Section: Literature Reviewmentioning

confidence: 97%

“…conducted HOCP using logistic regression (LR), k-nearest neighbor (KNN) and categorical boosting (CatBoost) and found that CatBoost can obtain higher prediction accuracy than LR and KNN Saputro and Nanang (2021). conducted HOCP using random forest (RF) and extra tree classifiers and identified 79% of all hospitality order cancellers Andriawan et al (2020). found that RF is more accurate in HOCP than XGBoost, CatBoost and light gradient boosting machine (LightGBM) in HOCP Satu et al (2020).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Hospitality order cancellation prediction from a profit-driven perspective

Liu

Jiang²,

Wang³

et al. 2022

IJCHM

View full text Add to dashboard Cite

Purpose This study/paper aims to reach the core objective of hospitality order cancellation prediction (HOCP), that is, to identify potential cancellers from many customer bases, thereby enhancing the effectiveness of customer retention campaigns. However, few studies have focused on predicting hospitality order cancellation. Design/methodology/approach A novel profit-driven model for predicting hospitality order cancellation is proposed to bridge this research gap. The authors construct profit-driven extreme gradient boosting (XGBoost) based on a grid search on HOCP to maximize profit by selecting optimal hyperparameters of XGBoost. Findings Real-world data set is analyzed, and the proposed model yields more profits than other predictive models. Sensitivity analysis proves that the proposed model is robust to the key hyperparameter and application scenario. Furthermore, some preventive measures based on visual analysis results are provided to reduce the cancelled probability of orders. Research limitations/implications This research will help hotel managers to transfer the modeling goal to profit orientation and encourage relevant researchers to interpret the prediction results of models for hotel order cancellation prediction in a post hoc manner. Besides, the proposed model can be applied to various enterprises with different average order profits and help managers optimize revenue management. Originality/value This research expands the relevant literature and offers guidance for predicting hospitality order cancellation from a profit-driven perspective at the customer level. The proposed model can provide macro-control to hotel managers and obtain the most satisfactory profits in micro-control.

show abstract

Section: Literature Reviewmentioning

confidence: 97%

mentioning

confidence: 99%

Hospitality order cancellation prediction from a profit-driven perspective

Liu

Jiang²,

Wang³

et al. 2022

IJCHM

View full text Add to dashboard Cite

show abstract

“…Furthermore, a study presented the use of a machine learning approach to predict the likelihood of a hotel booking cancellation using the CRISP-DM method. The method is useful for gaining a thorough understanding of the data and in assisting the hotel industry in determining which transactions will be cancelled soon [29]. The study proposed that "lead-time" would be an important variable that would play a role in booking cancellation.…”

Section: Research Backgroundmentioning

confidence: 99%

Implementation of Ensemble Machine Learning Techniques on Hotel Reservation System

Alotaibi¹

2023

IJETAE

View full text Add to dashboard Cite

The hotel industry effectively supports the wider national economy. The sector supports the nation's efforts to promote cultural events and improve visitor amenities. The primary issue this sector is currently dealing with is a hotel booking cancellation. Because of the impact this issue has on the hotel industry's resource allocation, labour needs, customers’ satisfaction, and overall decision-making process. In result, the hotel's reputation, business processes, and financial performance may also suffer. The major goal of this research is to create a model that will help the hotel industry make wise decisions. We completed the necessary data preprocessing and transformation procedures using the Kaggle hotel bookings dataset. Furthermore, we employed a number of machine learning methods to predict the cancellation requests in the future. Additionally, this study used a number of ensemble techniques, including voting, stacking, and bagging, to improve the model's accuracy. The findings showed that the stacking strategy outperformed all other models and had an accuracy rate of 86.76%. The proposed model and analysis discussed in this paper may help the hotel sector forecast the kinds of requests that will likely be cancelled in the future. Keywords—Hotel Booking Cancellation, Machine Learning, Ensemble Machine Learning, Classification

show abstract

“…During the Data Understanding stage, we collect and analyze data to identify information and assess its quality. This helps you figure out the percentage of data we'll analyze, including the amount to retrieve and the data that's stuck or running smoothly [7].…”

Section: Data Understandingmentioning

confidence: 99%

Implementation of the Decission Tree Algorithm to Determine Credit Worthiness

Abdussomad,

Kurniawan,

Wibowo

2023

Compiler

View full text Add to dashboard Cite

Credit is a loan from a bank that needs to be repaid with interest. In practice, problematic credit or bad credit often occurs due to less thorough credit analysis in the credit granting process, or from bad customers. This research aims to predict creditworthiness using the Decision Tree Classification Algorithm and find a solution for determining it. This research uses the CRISP-DM (Cross-Industry Standard Process for Data Mining) method. This research method tests the effects of using the decision tree, Support Vector Machine, and Naïve Bayes model with the Decision Tree Classification Algorithm. The decision tree classification algorithm accurately analyzed problem loans and non-problem debtors at 93.49%. The decision tree algorithm test results are better than the support vector machine by 3.45%, and naïve bayes by 13.03%. The results of our study were also 4.16% better than the previous study. This research has also implemented the selected model in the form of website application deployment.

show abstract

Prediction of Hotel Booking Cancellation using CRISP-DM

Cited by 7 publications

References 8 publications

Hospitality order cancellation prediction from a profit-driven perspective

Hospitality order cancellation prediction from a profit-driven perspective

Implementation of Ensemble Machine Learning Techniques on Hotel Reservation System

Implementation of the Decission Tree Algorithm to Determine Credit Worthiness

Contact Info

Product

Resources

About