Creating a campaign involves providing the basic properties for it; but keep in mind that you also need to configure other elements of a campaign in order to be able to execute analysis over the data.

If the campaign is already created, then you can check the following links, which may be more appropiate : 

First thing you need to provide is a name and description for your campaign.

Then you will need to set the start and end dates; these dates apply mostly to agent based executions (automatic executions).

In the case your company has the option to work with macrocampaigns, you can also choose here which macrocampaign this campaign belongs to.

Next step is to specify the type of target (binary or continous)

Finally you need to configure 

  • the Source you will be using in this campaign
  • test percentage, which is the percentage of rows you want to use for training or test. Data provided will be splitted in two different groups, test and train. Train data will be used to calculate the best model and test data will be used to validate it.
  • rows to use, nbt uses more than 50 different predictive models.This setting is used to select the best model, being the one which maximizes the gain over a random selection in these percentage of the data.

Once defined all these properties, you will be taken to the confirmation tab, where you can review all the settings (only when creating). 

Just click on Finish to create the campaign.