Scholarlink

<p>The hospitality industry significantly contributes to the economic growth of East Java, particularly in Surabaya. Hotel occupancy rates in Surabaya are notably influenced by factors such as urban density, local community lifestyle, and regional economic activity. Optimizing occupancy rates is crucial yet challenging, thus requiring advanced analytical approaches. This study aims to predict hotel occupancy rates using machine learning techniques, specifically by integrating Support Vector Regression (SVR) with Genetic Algorithms (GA). SVR, an advanced regression adaptation of the Support Vector Machine (SVM), is optimized through GA to enhance prediction accuracy. The combined SVR-GA method successfully minimized prediction error, achieving an impressively low Root Mean Square Error (RMSE) value of 0.0186. The integration of SVR and GA provides superior predictive performance, effectively capturing complex occupancy patterns specific to the hospitality industry in Surabaya. This methodological innovation underscores the potential for significant accuracy improvements compared to traditional forecasting methods. Practically, the proposed model offers hospitality managers in Surabaya a powerful predictive tool, enabling more informed decisions in occupancy management, strategic pricing, and revenue optimization.</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

Optimizing occupancy of hospitality sector using Support Vector Regression and Genetic Algorithm

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年6月28日

查看全文
  • Mohamad Yusak Anshori,
  • Teguh Herlambang,
  • Mohd Fathi Abu Yaziz

摘要

The hospitality industry significantly contributes to the economic growth of East Java, particularly in Surabaya. Hotel occupancy rates in Surabaya are notably influenced by factors such as urban density, local community lifestyle, and regional economic activity. Optimizing occupancy rates is crucial yet challenging, thus requiring advanced analytical approaches. This study aims to predict hotel occupancy rates using machine learning techniques, specifically by integrating Support Vector Regression (SVR) with Genetic Algorithms (GA). SVR, an advanced regression adaptation of the Support Vector Machine (SVM), is optimized through GA to enhance prediction accuracy. The combined SVR-GA method successfully minimized prediction error, achieving an impressively low Root Mean Square Error (RMSE) value of 0.0186. The integration of SVR and GA provides superior predictive performance, effectively capturing complex occupancy patterns specific to the hospitality industry in Surabaya. This methodological innovation underscores the potential for significant accuracy improvements compared to traditional forecasting methods. Practically, the proposed model offers hospitality managers in Surabaya a powerful predictive tool, enabling more informed decisions in occupancy management, strategic pricing, and revenue optimization.