How smart is e-tourism? A systematic review of smart tourism recommendation system applying data management

Abstract Extensive research has been conducted on e-tourism spanning a wide range of concepts, challenges and concerns discussed in tourism recommender systems (TRS). Smart tourism can be considered a logical progression from e-tourism laid with the extensive adoption of information and communication technologies and connecting the physical and digital worlds by taking advantage of 12 ‘smart key concepts’ such as privacy protection, Internet of Things and augmented reality, among others. Consequently, several disparate types of research have existed in various classes of TRS that have accomplished smart key concepts where others have failed. However, such piecemeal development is insufficient for a pragmatic smart tourism solution. Accordingly, the current study complements the academic literature with a systematic review that covers all main aspects of the e-tourism management system applied to the smart tourism concepts over the last eight years of publication. This study also provides a state-of-the-art e-tourism data management classification taxonomy based on smart concepts and reviews works in different fields against that classification. To this end, we reviewed the ScienceDirect, IEEE Xplore and Web of Science databases. A total of 1240 papers were collected from 2013 to 2020. The retrieved articles were filtered according to the defined inclusion criteria. Finally, 65 articles were selected and classified into two categories. The first category includes smart-based TRS that accounts for 87.70% (n = 57/65) and classified into four approaches: collaborative filtering, content model, context model and hybrid model. The second category includes tourism marketing that accounts for 12.30% (n = 8/65). This multi-field systematic review has exposed new research opportunities, motivations, recommendations and challenges and limitations that need attention for the synergistic smart integration of interdisciplinary studies. The reliability and acceptability of smart-based TRS approach from the implemented 12 smart key concepts show a significant difference. Analysis shows that the content model-based approach has a highly important effect on smart e-tourism, i.e., applying numerous smart key concepts in higher mean (40.2%). Results of several past studies that used a content model-based approach were nearly perceived as smart e-tourism. The smartly achieved key concepts for hybrid and context-based approaches have approximate means of (37.9%) and 36.6%, respectively, thereby confirming results. The results of tourism marketing and collaborative filtering approaches are worse than the previously reported results, achieving means of (33.3%) and (30.3%), respectively. This study is a useful guide for researchers and practitioners in providing avenues and valuable information for future research. This study is also expected to address the ambiguity of e-tourism and smart tourism trends.