CARMA Node Expert Options

For those with detailed knowledge of the CARMA node's operation, the following expert options allow you to fine-tune the model-building process. To access expert options, set the mode to Expert on the Expert tab.

Exclude rules with multiple consequents. Select to exclude “two-headed” consequents—that is, consequents that contain two items. For example, the rule bread & cheese & fish -> wine&fruit contains a two-headed consequent, wine&fruit. By default, such rules are included.

Set pruning value. To conserve memory, the CARMA algorithm used periodically removes (prunes) infrequent item sets from its list of potential item sets during processing. Select this option to adjust the frequency of pruning, and the number you specify determines the frequency of pruning. Enter a smaller value to decrease the memory requirements of the algorithm (but potentially increase the training time required), or enter a larger value to speed up training (but potentially increase memory requirements). The default value is 500.

Vary support. Select to increase efficiency by excluding infrequent item sets that seem to be frequent when they are included unevenly. This is achieved by starting with a higher support level and tapering it down to the level specified on the Model tab. Enter a value for Estimated number of transactions to specify how quickly the support level should be tapered.

Allow rules without antecedents. Select to allow rules that include only the consequent (item or item set). This is useful when you are interested in determining common items or item sets. For example, cannedveg is a single-item rule without an antecedent that indicates purchasing cannedveg is a common occurrence in the data. In some cases, you may want to include such rules if you are simply interested in the most confident predictions. This option is unselected by default.