Sample size is related to data mining. Data mining is available only when DB2 Intelligent Miner is installed.
When data mining runs it creates segmentation models based on only a sample of the data. Once a model has been built on the sample, it is applied to the whole population. If the sample is too large, your system's performance may degrade. You can reduce the sample size to improve performance but the accuracy of the mining results may also degrade.
To change the sample size, use the Get Random Sample function in DB2 Intelligent Miner for Data.
- Open DB2 Intelligent Miner for Data:
- From the Start menu, click Programs > IBM Intelligent Miner for Data V8.1 > Intelligent Miner.
- Expand Processing and select Get Random Sample. Parameters display in the right pane.
- Double-click a parameter to see its sampling rate in a new window.
- Change the sample size by entering a new value in the Sample Percentage field. Specify the sample size as a percentage of the input data.
- Click Apply.
- Click OK.
- Repeat steps 3 through 6 for each parameter you want to change.
For more information on the Get Random Sample function, refer to your DB2 Intelligent Miner for Data documentation.