In this task, you finish building the project ontology by creating a scorecard. The FraudCriteria scorecard will store the criteria used to determine fraud, not any specific data about customer accounts. In this example, you will use this scorecard in rules.
What is a scorecard? A scorecard is a special type of concept. A scorecard serves as a static variable. You can use a scorecard resource to track key performance indicators or any other information. Unlike concepts, there is only one instance of a scorecard. You create the scorecard at design time. Its values can be viewed and updated using rules.
It is more accurate to say there is one instance of a scorecard per inference agent. This tutorial uses one inference agent. However, in the next tutorial you will deploy multiple agents, and each has its own instance of the scorecard. This enables scorecards to be used for local purposes and minimizes contention between the agents. Do not use scorecards as a mechanism to share data between multiple agents.
You have now set up the ontology for the project, that is, the definitions of all the events, concepts, and scorecards that are needed to store information (facts) about possible fraud detection.