42
International Journal of Research in Business and Social Science
IJRBS Vol.3 No.4, 2014 ISSN: 2147-4478
available online at www.ssbfnet.com
Implementation of Chaid Algorithm: A Hotel Case
Celal Hakan Kağnicioğlu
a
, Mune Moğol
b
a
Ass.Prof. Anadolu University, Faculty of Economics and Administrative Sciences, Department of Business Administration, Eskişehir, 26170,
Turkey
b
Res. Ass., Anadolu University, Tourism Faculty, Department of Tourism Management, Eskişehir, 26170, Turkey
Abstract
Today, companies are planning their own activities depending on efficiency and effectiveness. In order to have plans for the future
activities they need historical data coming from outside and inside of the companies. However, this data is in huge amounts to
understand easily. Since, this huge amount of data creates complexity in business for many industries like hospitality industry,
reliable, accurate and fast access to this data is to be one of the greatest problems. Besides, management of this data is another big
problem. In order to analyze this huge amount of data, Data Mining (DM) tools, can be used effectively. In this study, after giving
brief definition about fundamentals of data mining, Chi Squared Automatic Interaction Detection (CHAID) algorithm, one of the
mostly used DM tool, will be introduced. By CHAID algorithm, the most used materials in room cleaning process and the relations
of these materials based on in a five star hotel data are tried to be determined. At the end of the analysis, it is seen that while some
variables have strong relation with the number of rooms cleaned in the hotel, the others have no or weak relation.
Keywords: Data Mining; CHAID; Tourism; Hotel
JEL: M31; M37; M39.
© 2014 Published by SSBFNET
1. Introduction
Most of the companies try to establish their own databases for prediction and further analysis. Data mining is the
collection of techniques which are discovered the hidden pattern of data from huge big datasets stored in databases.
Data mining is the process of understanding, preparing, modelling, and analyzing data. It is the way of extracting
potentially useful information and knowledge and uncovering unknown patterns and relationships hidden in datasets
(Rupnik, Kukar, Bajec, & Krisper, 2006; Ha & Park, 1998; Clifton & Thuraisingham, 2001; Shaw, Subramaniam,
Tan, & Welge, 2001; Huang, Wu, & Chou, 2013; Cheng & Chen, 2009; Gargano & Raggad, 1999; Windle, 2004;
Han, Kamber, & Pei, 2012: 2).