Applying Data Mining to Hotel Customer Segmentationand Marketing Mix Analysis

碩士 === 國立高雄第一科技大學 === 資訊管理所 === 94 === ABSTRACT Under the intense competition within the tourist industry, hotel hosts are actively studying CRM to attract and keep the most important customers. How to allot customers to fully utilize hotel resources and characteristics, design a customary tourist a...

Full description

Bibliographic Details
Main Authors: Kuo-sung Wang, 王國嵩
Other Authors: Cheng-lung Huang
Format: Others
Language:zh-TW
Published: 2006
Online Access:http://ndltd.ncl.edu.tw/handle/59175373117508455920
Description
Summary:碩士 === 國立高雄第一科技大學 === 資訊管理所 === 94 === ABSTRACT Under the intense competition within the tourist industry, hotel hosts are actively studying CRM to attract and keep the most important customers. How to allot customers to fully utilize hotel resources and characteristics, design a customary tourist accommodation product set, attract more consumers, and raise a customer’s satisfaction and returning rate are the hotel’s primary issues. A business systematic database can provide objective and helpful information and aid to partition customers and be used as the reference for making sales decisions. It helps to manage customer relationships and prompts hotel’s profit and competitiveness. This study case was taken from examples of the international tourist hotels—analyses and researches of customer profile data and 44,307 transaction records in hotels from 2002 to 2005. This study summated brand accepted the time length concept with RFM as the foundation to analyze customer’s value by using the data mining technique to cluster customers. With the clustered result, development for the classification model and the marketing combination assisted the sales department to select the first priority customers, to design the most proper products, and to undertake the best sales campaign. The results of this hotel study case shows that after the hotels adapted the improved RFMT model and utilized the clustering and classifying techniques of data mining, all customers were separated effectively and precisely, including emerging customers, low value customers, stagnated customers, and loyal customers. According to the customer data clustering result, the established classification database decision tree, and the marketing mix association analysis provided the referred information for decision makers to make a proper sales strategy.