engleski

Neural Network Modelling of Tourist Temporal Behavior

Many services and related consumer needs are temporally dependent: they oscillate and vary over time. This article emphasizes the vast potential of using neural networks in the research of tourist temporal behavior, a topic which has hardly been examined so far. Our approach is based on a comprehensive survey of customer sastisfaction and consumption administered in 12 coffee bars over 10 days, on a sample consisting of 931 customers. Along with the predictors and 14 dependent variables, the visit time was recorded, as well. The data was used to create different neural network models for predicting tourist satisfaction and consumption. The most valuable results were obtained from simulations using neural network models in which the following parameters were varied: different time periods (from 7 to 24 hours), days of the week, as well as other input parameters and aspects of service consumption in the aforementioned establishments, such as gender, age and the reasons for visiting those establishments). The results have also revealed the behaviour of different groups of customers in different establishments, competition between different service providers and rivalry between different groups of clients regarding using services within the same establishment at the same time. The knowledge thus obtained has been verified by a group of experts, familiar with the consumption in the establishments where the research was conducted. This paper has proved that it is possible to investigate, collect data and subsequently model temporal customer behaviour in a precise manner so as to gain a deepeer understanding of interdependencies. Possible applications of these models are: prediction of the impact of particular changes on customer satisfaction and money spent by particular clients, enhancement of business through adaptation and improvement of products and business practices of particular service providers, as well as their application in dynamic mobile recommendation systems for individual tourists.

neural networks modelling; behavior simulation; temporal tourists behavior; consumer satisfaction; temporal data mining

nije evidentirano