Scholarlink

<p>Surabaya, the capital of East Java, has transformed into a metropolitan center with significant prospects for tourism business expansion, principally fueled by its historical sites. The consistent rise in visitor arrivals underscores the urgent demand for sufficient lodging, driving swift expansion in the hotel sector. This advancement has been augmented by the extensive utilization of internet hotel booking programs, heightening competition inside the hotel industry. Hotel management are increasingly utilizing sophisticated, data-driven revenue forecasting algorithms to sustain competitive pricing and enhance profitability. This study evaluates the efficacy of two machine learning techniques—Gradient Boosting Regression (GBR) and Support Vector Regression (SVR)—in predicting hotel revenue. Empirical findings indicate that both models possess strong predictive ability, with Gradient Boosting Regression attaining a Root Mean Square Error (RMSE) of 0.044, marginally surpassing the Support Vector Regression technique, which yielded an RMSE of 0.055. These findings highlight the practical significance of advanced machine learning methodologies in improving revenue management tactics in the hospitality industry.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predicting hotel revenue using gradient boosting regression and support vector regression: a comparative analysis

  • 期刊论文

  • 发表日期：

    2025年9月4日

查看全文
  • Mohamad Yusak Anshori,
  • Puspandam Katias,
  • Teguh Herlambang,
  • Nur Shabrina Meutia,
  • Zuraini Binti Othman,
  • Mohd Sanusi Azmi

摘要

Surabaya, the capital of East Java, has transformed into a metropolitan center with significant prospects for tourism business expansion, principally fueled by its historical sites. The consistent rise in visitor arrivals underscores the urgent demand for sufficient lodging, driving swift expansion in the hotel sector. This advancement has been augmented by the extensive utilization of internet hotel booking programs, heightening competition inside the hotel industry. Hotel management are increasingly utilizing sophisticated, data-driven revenue forecasting algorithms to sustain competitive pricing and enhance profitability. This study evaluates the efficacy of two machine learning techniques—Gradient Boosting Regression (GBR) and Support Vector Regression (SVR)—in predicting hotel revenue. Empirical findings indicate that both models possess strong predictive ability, with Gradient Boosting Regression attaining a Root Mean Square Error (RMSE) of 0.044, marginally surpassing the Support Vector Regression technique, which yielded an RMSE of 0.055. These findings highlight the practical significance of advanced machine learning methodologies in improving revenue management tactics in the hospitality industry.