Analysis of the spatio-temporal pattern of restaurant performance using POI data and STFTiS model

Tourism as a distinguished 21st-century phenomenon requires capital flows, people, cultures and a persistent interaction between the aforementioned variables. The significance of tourism in this era depends mostly on its economic cycle; And high economic dynamism, both locally and internationally, has led to the expansion and development of the spectrum of services in this field. One of the pillars of this development is the use of public information, which is a very powerful and diverse dataset that can be used for free in spatial decision making. Among the spatial decision-making criteria in tourism is the analysis of datasets regarding restaurants, which can be used to measure their general performance in terms of service. Therefore, in this research, a series of public-opinion-based information in regards to restaurants were collected from different social networks; And the goal is to examine the quality of service for restaurants in different geographic units with the help of STFTiS model. In this regard, the information obtained from users' comments after preparation and sorting were processed with the help of clustering methods. Thereupon, after clustering results and the ratings given by users across all cities of Mazandaran province were integrated, by analysing the correlation between them and status of tourism in those cities, the results could be validated and policies regarding the development of restaurants across the province could be planned. The results obtained are that the cities of Ramsar, Nowshahr, Tankabon and Mahmudabad are cities with good performance in terms of tourism and services. Also, the similarity of 4 out of 5 cities that can be developed in the results of the proposed method compared to similar articles obtained from other methods shows an 80% match.