Publication: A qualitative attraction ranking model for personalized recommendations
Submitted Date
Received Date
Accepted Date
Issued Date
2018
Copyright Date
Announcement No.
Application No.
Patent No.
Valid Date
Resource Type
Edition
Resource Version
Language
en
File Type
No. of Pages/File Size
ISBN
ISSN
1757-9880
eISSN
Scopus ID
WOS ID
Pubmed ID
arXiv ID
item.page.harrt.identifier.callno
Other identifier(s)
Journal Title
Journal of Hospitality and Tourism Technology
Volume
9
Issue
1
Edition
Start Page
End Page
Access Rights
Access Status
Rights
Rights Holder(s)
Physical Location
Bibliographic Citation
Research Projects
Organizational Units
Authors
Journal Issue
Title
A qualitative attraction ranking model for personalized recommendations
Alternative Title(s)
Author(s)
Author’s Affiliation
Author's E-mail
Editor(s)
Editor’s Affiliation
Corresponding person(s)
Creator(s)
Compiler
Advisor(s)
Illustrator(s)
Applicant(s)
Inventor(s)
Issuer
Assignee
Other Contributor(s)
Series
Has Part
Abstract
Purpose This paper aims to find a way to personalize attraction recommendations for travelers. The research objective is to find a more accurate way to suggest new attractions to each traveler based on the opinions of other like-minded travelers and the traveler’s preferences. Design/methodology/approach To achieve the goal; developers have created a personalized system to generate attraction recommendations. The system considers an individual traveler’s preferences to construct a qualitative attraction ranking model. The new ranking model is the result of blending two processes: K-means clustering and the analytic hierarchy process (AHP). Findings The performance of the developed recommendation system has been assessed by measuring the accuracy and scalability of the ranking model of the system. The experimental results indicate that the ranking model always returns accurate results independent of the number of attractions and the number of travelers in each cluster. The ranking model has also proved to be scalable because the processing time is independent of the numbers of travelers. Additionally; the results reveal that the overall system usability is at a very satisfactory level. Research limitations/implications The main theoretical implication is that integrating the processes of K-means and AHP techniques enables a new qualitative ranking model for personalized recommendations that deliver only high-quality attractions. However; the designed recommendation system has some limitations. First; it is necessary to manually update information about the new tourist attractions. Second; the overall response time depends on the internet bandwidth and latency. Practical implications This research contributes to the tourism business and individual travelers by introducing an accurate and scalable way to suggest new attractions to each traveler. The potential benefit includes possible increased revenue for travel agencies that offer personalized package tours and support individual travelers to make the final travel decisions. The designed system could also integrate with itinerary planning systems to plot out a journey that pinpoints what travelers will most enjoy. Originality/value This research proposes a design and implementation of a personalized recommendation system based on the qualitative attraction ranking model introduced in this article. The novel ranking model is designed and developed by integrating K-means and AHP techniques; which has proved to be accurate and scalable.
研究目的 本研究主要探索如何建立个性化旅游胜地推荐模型。本研究通过分析旅游兴趣相似的游客意见和游客偏好选择; 建立一种更加准确推荐游客需要的旅游胜地方法。 研究设计/方法/途径 为了达到研究目的; 本研究建立了一种个性化推荐旅游胜地的信息系统。其系统通过分析每个游客的旅游偏好来建设一种定性旅游胜地排名模型。这种新型模型主要通过结合以下两种分析算法：（1）K平均聚类算法（K-means clustering）（2）层次分析法（AHP）。 研究结果 本研究建立的推荐信息系统经过了准确率和拓展性的测评。实验结果表明这种排名模型的准确率并不受旅游胜地多少和游客样本大小的影响。此外; 这种排名模型也具有拓展性; 因为算法时间并不受游客样本大小的影响。最后; 研究实验表明此排名模型客户体验性达到合格满意要求。 研究理论限制/意义 本研究的主要理论意义在于其结合了K平均聚类算法和层次分析法; 并建立了一种新型定性排名模型; 这种排名模型个性化地推荐更高质量的旅游胜地给游客。然而; 这种推荐信息系统有一些局限性。第一; 新旅游胜地的信息需要手动输入。第二; 整个系统的处理时间决定于网络带宽和延迟状况。 研究实践意义 本研究的实践意义在于其建立了一种准确和具有拓展性的新型旅游胜地推荐模型。这种模型的潜在价值将有利于旅游机构提供定制化旅游套餐和帮助游客制定旅游计划。此外; 这种模型还可以结合旅游路线计划系统以制定更加使游客满意的旅游行程。 研究原创性/价值 本研究推荐了一种基于定性旅游胜地排名模型的个性化旅游推荐模型。这种新型的排名模型结合K平均聚类算法和层次分析法; 实验证明这种模型更具准确性和拓展性。
研究目的 本研究主要探索如何建立个性化旅游胜地推荐模型。本研究通过分析旅游兴趣相似的游客意见和游客偏好选择; 建立一种更加准确推荐游客需要的旅游胜地方法。 研究设计/方法/途径 为了达到研究目的; 本研究建立了一种个性化推荐旅游胜地的信息系统。其系统通过分析每个游客的旅游偏好来建设一种定性旅游胜地排名模型。这种新型模型主要通过结合以下两种分析算法：（1）K平均聚类算法（K-means clustering）（2）层次分析法（AHP）。 研究结果 本研究建立的推荐信息系统经过了准确率和拓展性的测评。实验结果表明这种排名模型的准确率并不受旅游胜地多少和游客样本大小的影响。此外; 这种排名模型也具有拓展性; 因为算法时间并不受游客样本大小的影响。最后; 研究实验表明此排名模型客户体验性达到合格满意要求。 研究理论限制/意义 本研究的主要理论意义在于其结合了K平均聚类算法和层次分析法; 并建立了一种新型定性排名模型; 这种排名模型个性化地推荐更高质量的旅游胜地给游客。然而; 这种推荐信息系统有一些局限性。第一; 新旅游胜地的信息需要手动输入。第二; 整个系统的处理时间决定于网络带宽和延迟状况。 研究实践意义 本研究的实践意义在于其建立了一种准确和具有拓展性的新型旅游胜地推荐模型。这种模型的潜在价值将有利于旅游机构提供定制化旅游套餐和帮助游客制定旅游计划。此外; 这种模型还可以结合旅游路线计划系统以制定更加使游客满意的旅游行程。 研究原创性/价值 本研究推荐了一种基于定性旅游胜地排名模型的个性化旅游推荐模型。这种新型的排名模型结合K平均聚类算法和层次分析法; 实验证明这种模型更具准确性和拓展性。