Abstract
Traditional hospitality research on relationship modeling has concentrated on mathematical functions. A major restriction of mathematical function-based modeling techniques is their inability to handle non-numeric data. This article presents a new approach that applies rough set theory to model the relations within a set of numeric and non-numeric hotel expenditure data. The output of relationship modeling from a rough set approach is a set of decision rules. A hospitality information table was established, using data from the Hong Kong Tourist Association in the period 1983 to 1996. Among the 82 entries in the established information table, 65 were used to induce decision rules to model the data relation ship, and the remainder were used to test accuracy by comparing actual and forecasted hotel expenditures. Empirical results indicated that 75% of the classified forecasting samples fell into the same categories as their actual counterparts, and there was no significant difference between the actual and forecasted data.
Original language | English |
---|---|
Pages (from-to) | 359-375 |
Number of pages | 17 |
Journal | Journal of Hospitality and Tourism Research |
Volume | 22 |
Issue number | 4 |
DOIs | |
Publication status | Published - 1 Dec 1998 |
Keywords
- Forecasting
- Hong Kong
- Hotel expenditures
- Information table
- Rough set theory
ASJC Scopus subject areas
- Education
- Tourism, Leisure and Hospitality Management