Applying Data Mining to Loyal Customer Identification and Buying Behavior - The Case of Hypermarket
碩士 === 東海大學 === 管理碩士學程在職進修專班 === 95 === The tide of global economy, consumer buying behavior is changed and the number of channels is increased day by day in recent year. In the continuing economic depression, enterprises face a strong pressure in the limited market and competition. In the past, sel...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2007
|
Online Access: | http://ndltd.ncl.edu.tw/handle/27726361654944384166 |
id |
ndltd-TW-095THU00026002 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-095THU000260022016-05-27T04:18:21Z http://ndltd.ncl.edu.tw/handle/27726361654944384166 Applying Data Mining to Loyal Customer Identification and Buying Behavior - The Case of Hypermarket 資料採礦應用於忠誠顧客辨識及其購買行為模式之研究-以量販業為例 Meiji Hong 洪孟志 碩士 東海大學 管理碩士學程在職進修專班 95 The tide of global economy, consumer buying behavior is changed and the number of channels is increased day by day in recent year. In the continuing economic depression, enterprises face a strong pressure in the limited market and competition. In the past, sellers think that it could be made good sales if the quality of products was good enough, and customer, of course, would accept what supply from sellers. But now, good quality of products is the lowest request for keeping good customer relationship. To survive and maintain the competitive advantage, consumer-oriented strategy is a must for enterprises, find out the key customers and provide them a satisfied service. According to rule of 80/20, we know 80﹪of sales came from 20﹪of customers, and the customer loyalty and lose are the important target to judge the customer relation, therefore keeping a good customer relationship is very important to an enterprise. This study differs from the general sampling questionnaire method, but using data mining techniques to build up a business decision model. First stage, finding out the related important factor from customer loyalty and lose by using CROSS-TAB method. Second stage, adopting Cluster Analysis of Data Ming technique to divide customers to three groups and then test out the Discrimination effection by Discriminate Analysis. Finally, third stage , analyzing the buying behavior of the group of loyal customers according to history transaction record of customer database and provide a quick-effective decision supply model. The method is Using MANOVA to analyze the obvious differences : visiting ratio of buying time、purchasing ratio of variety product and price elasticity of promotion of different product. According to these information, making market segmentation and marketing strategy and then increasing customer satisfaction and loyalty. Hsu Sue-Ming 許書銘 2007 學位論文 ; thesis 101 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 東海大學 === 管理碩士學程在職進修專班 === 95 === The tide of global economy, consumer buying behavior is changed and the number of channels is increased day by day in recent year. In the continuing economic depression, enterprises face a strong pressure in the limited market and competition. In the past, sellers think that it could be made good sales if the quality of products was good enough, and customer, of course, would accept what supply from sellers. But now, good quality of products is the lowest request for keeping good customer relationship. To survive and maintain the competitive advantage, consumer-oriented strategy is a must for enterprises, find out the key customers and provide them a satisfied service.
According to rule of 80/20, we know 80﹪of sales came from 20﹪of customers, and the customer loyalty and lose are the important target to judge the customer relation, therefore keeping a good customer relationship is very important to an enterprise. This study differs from the general sampling questionnaire method, but using data mining techniques to build up a business decision model. First stage, finding out the related important factor from customer loyalty and lose by using CROSS-TAB method. Second stage, adopting Cluster Analysis of Data Ming technique to divide customers to three groups and then test out the Discrimination effection by Discriminate Analysis. Finally, third stage , analyzing the buying behavior of the group of loyal customers according to history transaction record of customer database and provide a quick-effective decision supply model. The method is Using MANOVA to analyze the obvious differences : visiting ratio of buying time、purchasing ratio of variety product and price elasticity of promotion of different product. According to these information, making market segmentation and marketing strategy and then increasing customer satisfaction and loyalty.
|
author2 |
Hsu Sue-Ming |
author_facet |
Hsu Sue-Ming Meiji Hong 洪孟志 |
author |
Meiji Hong 洪孟志 |
spellingShingle |
Meiji Hong 洪孟志 Applying Data Mining to Loyal Customer Identification and Buying Behavior - The Case of Hypermarket |
author_sort |
Meiji Hong |
title |
Applying Data Mining to Loyal Customer Identification and Buying Behavior - The Case of Hypermarket |
title_short |
Applying Data Mining to Loyal Customer Identification and Buying Behavior - The Case of Hypermarket |
title_full |
Applying Data Mining to Loyal Customer Identification and Buying Behavior - The Case of Hypermarket |
title_fullStr |
Applying Data Mining to Loyal Customer Identification and Buying Behavior - The Case of Hypermarket |
title_full_unstemmed |
Applying Data Mining to Loyal Customer Identification and Buying Behavior - The Case of Hypermarket |
title_sort |
applying data mining to loyal customer identification and buying behavior - the case of hypermarket |
publishDate |
2007 |
url |
http://ndltd.ncl.edu.tw/handle/27726361654944384166 |
work_keys_str_mv |
AT meijihong applyingdataminingtoloyalcustomeridentificationandbuyingbehaviorthecaseofhypermarket AT hóngmèngzhì applyingdataminingtoloyalcustomeridentificationandbuyingbehaviorthecaseofhypermarket AT meijihong zīliàocǎikuàngyīngyòngyúzhōngchénggùkèbiànshíjíqígòumǎixíngwèimóshìzhīyánjiūyǐliàngfànyèwèilì AT hóngmèngzhì zīliàocǎikuàngyīngyòngyúzhōngchénggùkèbiànshíjíqígòumǎixíngwèimóshìzhīyánjiūyǐliàngfànyèwèilì |
_version_ |
1718283139067936768 |