Knowledge discovery and computerized reasoning to assist tourist destination marketing

Abstract

In this paper we present a knowledge-based approach where data analysis and reasoning technologies are consolidated in order to provide solutions to common marketing planning problems. The aim was to support marketers in the tourist domain with intelligent electronic tools, which are designed to extract knowledge from survey data, maintain it in a suitable Knowledge Base and offer it for problem solving in a computerized environment to users who are not experts in analysis. The focus was to capture the image of Thessaloniki as tourist destination, to study the characteristics and needs of its visitors and to suggest important factors for improving the promotion of the city to individual visitor groups. The data sources were questionnaire-based primary surveys on the destination image and the expectations of visitors regarding their hotel. Multidimensional Multi-factorial Data Analysis was used as a knowledge extraction method to discover factors, associations and other hidden patterns that explain the behavior of tourists, as well as clusters that can be studied individually. An ontology-based Knowledge Modeling process was then developed to express the results of data analysis as machine understandable rules, so that an inference engine can respond to high-level queries. Special types of visitors were identified, such as visitors for nightlife and visitors for culture, and rules were formulated to predict the requirements of such types, in relation to parameters such as age and country. A query mechanism running on rule-based reasoning was successful in providing suggestions to typical marketing problems.