Profit Chart (Analysis Services - Data Mining)
You can view two types of charts in the Lift Chart tab of the Mining Accuracy Chart tab of Data Mining Designer: a lift chart and a profit chart. After you have configured the model and the data source to use, you can select the type of chart you want. After you select Profit Chart from the list, the Profit Chart Setting dialog box automatically opens. If you set the parameters that define a profit chart, the chart that is displayed in Mining Accuracy Chart tab automatically changes to a profit chart.
A profit chart displays the estimated profit increase that is associated with using a mining model. For example, if your model predicts which customers a company should contact in a business scenario, the profit chart incorporates information about the cost of conducting the targeted mailing campaign to contact x number of customers, and calculates the estimated profit. A typical profit chart shows an increase in profits up to a point, after which profits decrease as more of the population is contacted.
For example, in the Basic Data Mining Tutorial, you create a decision tree model, TM_Decision Tree, that predicts which AdventureWorks prospective customers are more likely to buy a bike. To generate a profit chart that represents the costs and benefits of sending a mailing to just those customers, you follow the general steps used to build a lift chart, and then configure the settings that are unique to profit charts. When you finish setting the chart parameters, the chart is automatically changed to a profit chart. The following diagram is based on these assumptions:
Which model should you use?
What value should be predicted?
[Bike Buyer] =1, meaning customers who are likely to buy a bike
What data set should be used to assess accuracy?
To assess accuracy and potential profit, you will use the test set that was saved when the mining structure was created.
When you actually create the mailing, you will use a different data set/
What is the total target population?
Out of all customers in the database, you will send targeted mailings to only 20,000.
What is the one-time cost of setting up a targeted mailing campaign for 20,000 people?
What is the per-unit cost for the targeted mailing campaign?
This amount will be multiplied by a number less than or equal to 20,000, depending on how many customers the model predicts are good candidates.
What profit or income can be expected from a successful result?
This amount will be used to project the total profit associated with high probability cases.
The Y-axis of the chart represents the profit, while the X-axis represents the percentage of the population that the company contacted.
The profit chart contains a gray vertical line that marks a percentage of the target population. You can move the line by clicking a location in the chart. Each time you move the line, the Mining Legend is updated to display the percentage value, a profit score, and the predict probability that is associated with the population percentage at the vertical gray line. If you move the gray line to the point in the chart where profits are the highest, you can use the predict probability value to determine a strategy for contacting customers.
By experimenting with this graph, you might determine that the peak of the profit curve is at 55 percent of the population and the associated predict probability is 20 percent. These results indicate that to achieve maximum profits you should only contact those customers whose response is predicted with a 20 percent or greater chance.
To create a profit chart, you follow these basic steps:
On the Input Selection, choose a model or models.
Choose a predictable attribute.
Optionally, specify a value to predict.
Choose the data source to use for evaluation
Change to chart view by clicking Lift Chart.
On the Lift Chart tab, select the type of chart you want by clicking the Chart Type list.
Configure options specific to profit charts.
For a step-by-step explanation of how to create all chart types, see How to: Create an Accuracy Chart for a Mining Model. The Basic Data Mining tutorial also contains a walkthrough of how to create a lift chart. For more information, see Testing Accuracy with Lift Charts (Basic Data Mining Tutorial).
The following list describes the parameters that you can set in the Profit Chart Settings dialog box.