Presenting the Prediction Model for the Return of Health Tourists based on Lifestyle Clusters (with Emphasis on Ramsar Hot Springs)

The aim of this study is to draw a prediction model for the return of spa tourist based on lifestyle clusters. This was done in two related steps. In the first step, Abdi and Kajbaf lifestyle questionnaire was used, which its reliability and validity were verified through Cronbach's alpha and first and second confirmatory factor analysis respectively. The views of 391 tourists were gathered on 10 dimensions of lifestyle. Then using cluster analysis, K mean obtained and based on Davis-Bouldin index, optimum cluster were acquired. Spa tourists were classified in three clusters. In the second step, using profound semi-structured interviews with tourists, 14 effective indicators were identified for their re-travel. Using Decision tree method, the decision models of return of tourists designed and the rules if-then related to each cluster were extracted. The results showed that the tourists in the first cluster are highly sensitive to travel costs such as the cost of accommodation and the cost of the spring, for the tourists in the second cluster, the health factors of the spring, security facilities, and employees’ behavior are the priorities for decision making. Finally, in the pattern of the decision of tourists in the third cluster, three indicators of spring health, accommodation facilities, and counseling and medical facilities are of great importance