DESIGN OF A MULTILEVEL INTELLIGENT DECISION SUPPORT SYSTEM FOR THE IMPROVEMENT OF TOURIST SATISFACTION

Knowledge discovery in databases (KDD) was used on survey data relating to tourists' attitudes and consumption in Croatia. The results of the KDD were exploited to develop 73 models of tourist satisfaction that achieved relatively high precision (average deviation of 6.4%). These models were then applied to simulate the ability to improve the total satisfaction of tourists by choosing a destination that was more appropriate to their preferences. The last simulation indicated that for 90% of tourists the degree of satisfaction could be improved by 25% to 160%. Finally, a second-generation intelligent system was conceptualized that uses the knowledge of tourist satisfaction models for decision making at the levels of: (a) a multimedia intelligent tourist advisor, (b) a destination management system, and (c) a system for optimizing national tourism promotion activities.