Article: Development of optimal tourist destination recommendation model using clustering and shortest-distance digitalisation Journal: International Journal of Work Innovation (IJWI) 2026 Vol.7 No.4 pp.522 - 542 Abstract: Tourism is a strategic sector that contributes significantly to economic growth. However, the challenge of determining tourist destinations that match preferences, location, and travel efficiency remains a barrier for tourists. This study aims to develop the best tourist destination recommendation system by integrating three main algorithms: K-means clustering to group destinations based on tourism potential, Floyd-Warshall to calculate the shortest distance between destinations, and weighted product for multi-criteria decision making based on attributes such as facilities, cost, and safety. The results show that the integration of the three algorithms can produce more accurate and efficient recommendations compared to a single approach. This system not only supports personalised travel recommendations but also contributes to smarter tourism digitalisation. Further development is recommended by adding real-time data, utilising machine learning to understand user preferences more deeply, and testing on a larger scale. Inderscience Publishers - linking academia, business and industry through research

Title: Development of optimal tourist destination recommendation model using clustering and shortest-distance digitalisation

Authors: Muqorobin; Tira Nur Fitria; Annisa Indah Mutiasari; Mahameru Rosy Rochmatullah

Addresses: Department of Informatics, Faculty of Technology, Institut Teknologi Bisnis AAS Indonesia, Indonesia ' Department of Islamic Economics, Faculty of Business Economics, Institut Teknologi Bisnis AAS Indonesia, Indonesia ' Department of Business Administration, Faculty of Social, Humanities and Arts, Universitas Sahid Surakarta, Indonesia ' Department of Accounting, Faculty of Business Economics, Universitas Muhammadiyah Surakarta, Indonesia

Abstract: Tourism is a strategic sector that contributes significantly to economic growth. However, the challenge of determining tourist destinations that match preferences, location, and travel efficiency remains a barrier for tourists. This study aims to develop the best tourist destination recommendation system by integrating three main algorithms: K-means clustering to group destinations based on tourism potential, Floyd-Warshall to calculate the shortest distance between destinations, and weighted product for multi-criteria decision making based on attributes such as facilities, cost, and safety. The results show that the integration of the three algorithms can produce more accurate and efficient recommendations compared to a single approach. This system not only supports personalised travel recommendations but also contributes to smarter tourism digitalisation. Further development is recommended by adding real-time data, utilising machine learning to understand user preferences more deeply, and testing on a larger scale.

Keywords: recommendation system; K-means clustering; Floyd-Warshall; weighted product; digitalisation of tourism.

DOI: 10.1504/IJWI.2026.157136

International Journal of Work Innovation, 2026 Vol.7 No.4, pp.522 - 542

Received: 23 Jan 2026
Accepted: 07 Apr 2026
Published online: 20 Oct 2026 *

Title: Development of optimal tourist destination recommendation model using clustering and shortest-distance digitalisation

Keep up-to-date