Creating the solution outline
Let us assume that at this moment we know nothing about who will respond to our marketing folders. In traditional marketing, we can provide discount coupons in the campaign folders with a barcode as a means to uniquely identify a customer. When a customer presents a discount coupon during the purchase then we know that the customer has responded to our marketing campaign. We can join the barcode on the discount coupon with the customer master data available in the company to find out who has responded to our campaign.
We will solve the problem of Furnitica using classification. Classification is a supervised learning method which uses historical data and past outcomes to predict future outcomes.
As an example, for a credit card company, the historical credit card approval data is shown in Table 1 :
|
Gender |
Age |
Owns a House |
Owns a Car |
Annual Salary in EUR |
Result |
|
M |
23 |
N |
N |
24000 |
Not Approved |
|
F |
35 |
Y |
N |
55000 |
Approved |
|
M |
40 |
Y |
Y |
52000... |
