Using Data Mining to Study Customer’s Consumption Pattern and Value

• Main Authors: Feng-Ching Lin, 林峰慶
• Other Authors: 黃信豪
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/x845h2

碩士 === 國立虎尾科技大學 === 工業工程與管理研究所在職專班 === 99 === The objective of this study has two folds: one is to analyze current customer’s consumption pattern of a telecommunication company’s alliance shop; the other is to perform customer classification based on the customer value. There are 1025 customer consumption records collected in the period from April to October, 2010. The PolyAnalyst 6.0 data mining software is used to analyze these record in this study. The results of this study show that: 1.51.71% of the customers choose monthly phone rate of NT$ 401~600. 2.60.39% of the customers do not apply for wireless internet service which means there are a lot of marketing opportunities in this area for the telecommunication company. 3.Customer applies for new service has the tendency towards not having wireless internet service. 4.Customer applies for extension service has the tendency towards selecting monthly internet rate of NT$ 100. 5.New customers have the tendency towards selecting mobile phone with the price of NT$ 2290, and then NT$ 3290. 6.Old customers have the tendency towards selecting mobile phone with the price less than NT$ 1490. 7.Customers are categorized into four classes: Class A customers are those who are paying higher monthly phone, having internet service, and using high-price mobile phone; Class B customers are those who are paying higher monthly phone rate, having internet service, and using mid-price mobile phone; Class C customers are those who are paying higher monthly phone rate, having no internet service, and using high-price mobile phone; Class D customers are those who are paying lower monthly phone rate, having internet service, and using high-price mobile phone. The numbers of customers classified in Class A, B, C, and D are 8, 28, 2, and 72, respectively.