Artificial Neural Network-Based Applications in Travel and Tourism Research: A Review and Case Study

Although frequently referred to as “black boxes”, artificial neural networks (ANN) find increasing application in intelligent- and recommender systems in a wide range of industries. In travel and tourism research ANNs have, however, not been extensively used so far. This is despite the fact that first empirical studies in peer-reviewed tourism journals have already been published in the late nineties, introducing ANNs as a valid alternative to traditional regression-based approaches, mostly with regard to demand forecasting purposes.
The aim of the present working-paper, on the one hand, is to provide an overview of available ANN-based studies published in top international travel and tourism journals, which may serve as a starting point and reference list for tourism researchers unfamiliar with ANNs. Basic concepts, main areas of application, as well as major advantages and disadvantages of ANN-based approaches, compared to traditional approaches, are highlighted.
On the other hand, this study aims to demonstrate the particular advantages and shortcomings of ANN-based applications using an empirical case example. For this purpose, this study uses data from a survey on attitudes and expenditures of tourists in Sarajevo, Bosnia and Herzegovina, conducted by the Institute for Tourism, Zagreb during summer 2010. In particular, a multilayer perceptron-based key-driver analysis is performed on the data to obtain insight into those destination attributes that have a predominant influence on the overall tourist experience in Sarajevo. Finally, results from the ANN-based analysis are opposed to results from a (traditional) regression-based key-driver analysis in order to identify possible significant differences between the approaches.

Kresic_Mikulic_Kozic_CTS_2013_Working_paper.pdf download View | Download
Keywords: Artificial Neural Network, Tourism Research
Categories: Tourism and its potential as a social force