About the Sample process in interactive flowcharts
The Sample process in interactive flowcharts fills the same purpose of Sample processes in batch flowcharts, to create one or more cells for different treatments, control groups, or a subset of data for modeling. However, the interactive flowchart Sample process contains only a subset of the features in the Sample process in batch flowcharts. The Sample Size Calculator and Maximum Cell Size are removed. Interactive flowchart Sample processes also have a new method for selecting the sample, the Deterministic Hash Function.
In a batch flowchart, the Sample process randomly selects some number of customers to be in a sample cell. In an interactive flowchart, however, during runtime, the Sample process works with a single customer at a time. When you define the % for a sample, you are defining the probability that a visitor becomes a member of that cell.
When defining sample cells, you should create at least two. Define a % for one sample, and select All Remaining for the other. If you do not do this, you will have undefined results. For example, if you create one 30% sample only, the remaining 70% of visitors are not assigned to any cell.
You must connect the Sample process to a PopulateSeg process to complete creating a smart segment you use for sampling.
The Deterministic Hash Function assigns visitors to samples randomly. However, if the same visitor passes through the interactive flowchart more than once, the visitor is placed in the same cell. If it is desirable, to ensure that visitors are eligible to be sampled differently after a period of repeatedly being placed in to the same cell, you should configure a reset time and periodicity for the deterministic hash function.
If you are creating a flowchart that copies the functionality of another flowchart, you can ensure that the Sample process uses the same Deterministic Hash Function as the original interactive flowchart by using the Hash Seed field. To place visitors in the same sample group, use the same number of cells, Hash Seed, Date of First Reset, and reset period in both Sample processes. Interact uses the value of the Hash Input seed and the reset date to determine which cell the visitor will be placed in.
To configure the Sample process
To configure the Sample process
1.
In an interactive flowchart in Edit mode, add a Sample process to the flowchart workspace.
2.
Provide input to the Sample process by connecting it to one or more data manipulation processes.
Data manipulation processes include Decision, Sample, or Select. You can connect the Sample process to an Interaction process also.
3.
Double-click the Sample process.
4.
Select one or more source cells from the Input drop-down list.
All output cells from any process connected to the Sample process are listed in the drop-down list. To use more than one source cell, select the Multiple Cells option. If more than one source cell is selected, the same sampling is performed on each source cell.
5.
Determine the number of samples that you want to create for each of your input cells, and enter that number in the # of Samples/Output Cells field.
By default, 3 samples are created for each input cell, with default names "Sample1", "Sample2" and "Sample3."
6.
Select each sample under the Output Name column and do the following:
The Edit Output Cell section is enabled.
a.
Place your cursor in the Cell Name text box and type to modify the sample name.
You can use any combination of letters, numbers, and spaces. Do not use any periods (.) or slashes (/ or \).
b.
Enter the probability that a visitor is assigned to the sample in the % field, or select the All Remaining checkbox.
The % must be less than 100.00.
7.
(Optional) Change the Hash Seed under the Deterministic Hash Function.
You need to change the Hash Seed only if you want this Sample process to assign visitors to cells the same way as another Sample process. The Hash Seed must be an alphanumeric value.
8.
(Optional) Enter the Date of first reset under the Deterministic Hash Function.
Click the ellipsis button Ellipsis button for a calendar to select the date.
9.
(Optional) Enter the number of days between resets.
10.
(Optional) Click the General tab to assign a name and notes to the process.
The name appears on the process in the flowchart. The notes appear when you mouse of the process in a flowchart.
11.
Click OK.
The process is configured and appears enabled in the flowchart.
You must connect the Sample process to a PopulateSeg process to complete creating a smart segment you use for sampling.
Copyright IBM Corporation 2015. All Rights Reserved.