Rough Set Analysis and Short-Medium Term Tourist Services Demand Forecasting

Rough Set Analysis and Short-Medium Term Tourist Services Demand Forecasting

Emilio Celotto (Department of Management, Ca' Foscari University of Venice, Italy), Andrea Ellero (Department of Economics and SSE, Ca' Foscari University of Venice, Italy) and Paola Ferretti (Department of Economics and SSE, Ca' Foscari University of Venice, Italy)
DOI: 10.4018/978-1-4666-4490-8.ch031
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Along with a growing interest in tourism research is the effort to establish innovative methodologies that are useful to guide the tourist operators and the policy makers in selecting forecasting techniques. Nevertheless, predicting tourist demand is still lacking at a microeconomic level, while it has become a flourishing theme of research uniquely at a macroeconomic level. The main goal is to analyze Italian tourists' behaviours on the basis of statistical surveys on households, life conditions, incomes, consumptions, travels, and vacation. This research is set in the framework of Rough Sets Theory, a Data Mining technique that can easily manage categorical variables. Hence, it is suitable for the exploitation of databases collecting sample surveys data. A large selection of variables from database Sinottica, containing information on social, cultural, and behavioural trends in Italy collected by means of a psychographic survey is provided by a leading market research organization, GfK Eurisko. By defining some decision rules, some interesting relations between consumer behaviours and their corresponding tourism choices are obtained.
Chapter Preview
Top

Rough Sets Theory: Some Basic Definitions

Rough Sets theory has been presented as a new mathematical tool for imperfect data analysis in (Pawlak, 1991). It concerns approximated knowledge of data that are collected in data table by means of two alternative approaches: lower/upper approximations of a data set and decision rules. In fact, Pawlak wrote: “Approximations and decision rules are two different methods to express properties of data. Approximations suit better to topological properties of data, whereas decision rules describe in a simple way patterns in data and are therefore the best means to communicate the results to the operators” (Pawlak, 2002).

Complete Chapter List

Search this Book:
Reset