Datamining group final version
Data Mining
Turn in your answers, along with excerpted output as relevant.
In this assignment, we will explore RFM segmentation, a technique used to group customers according to their aggregate purchase history with a company. We specifically look at how recently customers have purchased (R – recency), how often they have purchased
(F – frequency), and how much they have spent (M – monetary). We will then consider how these different segments responded to the offer to buy “The Art History of Florence.”
Use the CUSTOMER.SAV dataset for all parts of this assignment.
Team members:
WANG MAOXIN 53869782
LIU YI 53699531
YANG RONGRONG 53719976
HONG YIHAN 53735560
HUXIAO 53872424
MA CHUI YU 53916610
CHING KAM TO 53580292
1. Frequency Monetary of purchase
a) Create a bar graph plotting response rate summarized by total number of purchases (HINT: the response rate is equal to the mean of the BUYERS variable).
b) Does purchase frequency help to predict the likelihood that a customer responded to the offer to purchase “The Art History of Florence?” Explain your reasoning.
Yes. In the bar graph above, it can be seen generally that high purchase relate to high mean of bought “The Art History of Florence”. The person whose frequency is high is more likely to purchase “The Art History of Florence”.
c) Create a bar graph summarizing average dollars spent (including book and non-book items) by total number of purchases
d) How (if at all) does purchase frequency appear to be related to dollars spent and months since last purchase? Explain why any observed relationships might exist.
For b, d the command is Graphs/Bar/Simple. For (b) put BUYER in the Variable and make sure the “Other Statistic (Mean) is clicked.
From the bar graph above, it is obviously seen that Purchase frequency is positively related to dollars spent. There might be a strong correlation between purchase frequency and dollars spent. People who purchase more
Turn in your answers, along with excerpted output as relevant.
In this assignment, we will explore RFM segmentation, a technique used to group customers according to their aggregate purchase history with a company. We specifically look at how recently customers have purchased (R – recency), how often they have purchased
(F – frequency), and how much they have spent (M – monetary). We will then consider how these different segments responded to the offer to buy “The Art History of Florence.”
Use the CUSTOMER.SAV dataset for all parts of this assignment.
Team members:
WANG MAOXIN 53869782
LIU YI 53699531
YANG RONGRONG 53719976
HONG YIHAN 53735560
HUXIAO 53872424
MA CHUI YU 53916610
CHING KAM TO 53580292
1. Frequency Monetary of purchase
a) Create a bar graph plotting response rate summarized by total number of purchases (HINT: the response rate is equal to the mean of the BUYERS variable).
b) Does purchase frequency help to predict the likelihood that a customer responded to the offer to purchase “The Art History of Florence?” Explain your reasoning.
Yes. In the bar graph above, it can be seen generally that high purchase relate to high mean of bought “The Art History of Florence”. The person whose frequency is high is more likely to purchase “The Art History of Florence”.
c) Create a bar graph summarizing average dollars spent (including book and non-book items) by total number of purchases
d) How (if at all) does purchase frequency appear to be related to dollars spent and months since last purchase? Explain why any observed relationships might exist.
For b, d the command is Graphs/Bar/Simple. For (b) put BUYER in the Variable and make sure the “Other Statistic (Mean) is clicked.
From the bar graph above, it is obviously seen that Purchase frequency is positively related to dollars spent. There might be a strong correlation between purchase frequency and dollars spent. People who purchase more