The CARMA node uses an association rules discovery algorithm to discover association rules in the data.

Association rules are statements in the form:

if  antecedent(s)  then  consequent(s)

For example, if a Web customer purchases a wireless card and a high-end wireless router, the customer is also likely to purchase a wireless music server if offered. The CARMA model extracts a set of rules from the data without requiring you to specify input or target fields. This means that the rules generated can be used for a wider variety of applications. For example, you can use rules generated by this node to find a list of products or services (antecedents) whose consequent is the item that you want to promote this holiday season. Using watsonx.ai, you can determine which clients have purchased the antecedent products and construct a marketing campaign designed to promote the consequent product.

Requirements. In contrast to Apriori, the CARMA node does not require Input or Target fields. This is integral to the way the algorithm works and is equivalent to building an Apriori model with all fields set to Both. You can constrain which items are listed only as antecedents or consequents by filtering the model after it is built. For example, you can use the model browser to find a list of products or services (antecedents) whose consequent is the item that you want to promote this holiday season.

To create a CARMA rule set, you need to specify an ID field and one or more content fields. The ID field can have any role or measurement level. Fields with the role None are ignored. Field types must be fully instantiated before executing the node. Like Apriori, data may be in tabular or transactional format.

Strengths. The CARMA node is based on the CARMA association rules algorithm. In contrast to Apriori, the CARMA node offers build settings for rule support (support for both antecedent and consequent) rather than antecedent support. CARMA also allows rules with multiple consequents. Like Apriori, models generated by a CARMA node can be inserted into a data stream to create predictions.