Constraint-Based Recommendation in Tourism: A Multiperspective Case Study

Abstract

In many business-to-consumer (B2C) e-commerce scenarios, recommender systems (RS) have been shown to be valuable tools both for the online customer and the merchant. Such systems help customers find interesting items in large product assortments, increasing the chance of immediate online purchases and fostering long-term customer loyalty. However, standard technologies from classical RS application domains such as books and movies cannot be directly adopted in the tourism domain. This article presents a case study of a constraint-based RS that was integrated into a travel advisory system for an Austrian spa resort. The study analyzes the system and its environment from three perspectives. First, technological aspects of system development and maintenance are discussed; second, corresponding to the supplier’s view, the end user’s perspective is analyzed based on the findings of a study of the system’s usability and the perceived customer utility. Finally, the effectiveness of the system’s ability to positively affect user behavior is evaluated and discussed. The findings show that constraint-based RS not only help positively influence tourist behavior, but such systems can be built cost-effectively when using appropriate knowledge acquisition and maintenance tools.