|
|
The Caravan Case: Data Design
| Name |
The Caravan Case: Data Design |
| Description |
|
Figure 2
From the company database, a subset is extracted. Figure 2
gives an overview of the design of the company database. The database kernel
(grey area)
contains information on the customer in terms of policy buying behavior.
Purchasing additional information from the Mosiac information service routinely
extends this database. The learn-set consists of records that reflect the
relationship with the client. The two main groups of fields
are personal data (
mosaic region and customer
typing) and sales data
(data on sales behavior of the customer).
|
| |
|
|
|
Actual client behavior (what is the payment discipline, what is
the claim frequency), although potentially informative, is left out this data set.
It has already been used in selecting clients for this marketing
campaign: clients with commercially unsatisfactory behavior will not be seen as a
marketing target anyway.
|
| |
|
|
|
When viewing the data for individual clients, it appears that the
data as represented in the data set is a rough representation of the actual
client, where the client descriptions are educated guesses. Data for two types
of clients (a Successful hedonist and
a
Retired & religious client) are included.
Using this data, we want to predict which customers have the highest probability
of buying a caravan policy.
|
| |
|
|
|
| Case Study |
The Caravan Policy Case
|
|
|
