Policy Modeling Concepts

The key concepts for Policy Modeling are given in Table 1 below.

Table 1. Key concepts in Policy Modeling

Concept	Description
Projects	Policy Modeling projects manage the rule documents, the data model and the interview definition that make up a policy model.
Rules	Rules specify the logical relations between attributes and provide a framework against which attributes can be tested to determine their value. Rules describe relationships in terms of if-then logic.
Data model	The data model in Policy Modeling consists of the entities, relationships and attributes that make up the policy model.
Attributes	An attribute is a unit of data or fact. Attributes can either have a true or false value or take on a variable value (for example, a number, date, text and so on).
Entities	An entity represents a group of things (for example, people, objects, concepts) with rules or data in common.
Relationships	A relationship defines how two entities relate to each other so that rules can be written that reason from one entity to another.
Data mapping	The connection of Intelligent Advisor entities and attributes to tables and fields in a connected application.
Interviews	An interview is a web-based application which collects information from a user to find values for policy model goals.
Screens	A screen is what is displayed to the user during an interview. Screens have headings and can contain question text and answer fields, informative text, explanations, links to forms and more. Defining screens allows you to group attributes onto single views and provide the user with a sense of context by adding labels and headings to those screens.
Explanations	An explanation is a tailored list of reasons about how the value of the goal attribute was determined. It can be added to an interview or form to give the reasons for a particular decision. It is most suitable for displaying to customers and agents.
Checkpoints	Checkpoints are points in an interview where it can be saved, resumed and reviewed from. A checkpoint works by saving the progress of the interview, enabling interview users to easily resume an interview that they or someone else left partially completed. Checkpoints also allow completed interviews to be reviewed at any time.
Debugging	Debugging is the process of testing the logic of your rules to confirm that inputs match expected decisions.
Test Cases	A test case is one of the testing tools that can be used with Policy Modeling for repeated extensive testing and what-if analysis. Test cases specify the data that would be loaded at the start or entered during the interview and expected outcomes. Test cases can also be used with sample data (contained in a test data connector file) that specifies the data that would be dynamically loaded during an interview.
Hub	Intelligent Advisor Hub is a cloud-based console used to manage, share and deploy Policy Modeling, decision service and flow projects. It is also used to configure user permissions, connections and environments, and provides source control for collaboratively-authored projects.
Connections	Connections are used to retrieve metadata from external applications (such as Oracle Fusion, Oracle Integration or web service applications) via Intelligent Advisor Hub, and are how data is saved and loaded by interviews at runtime.
Users	Intelligent Advisor Hub user roles are needed for managing project versions and deployment, and for administering other users.
Deployment	Deployment is the first step in the process of making a project version available for Intelligent Advisor interview and web service interactions (the other step is activation). Deploying a project is performed in Policy Modeling. Deployment is managed via an Intelligent Advisor Hub.
Activation	Activation is the second step in the process of making a project version 'go live' for end users (the other step is deployment). Activation can be performed within Policy Modeling at the time of deployment, or later from Intelligent Advisor Hub. End users can then interact with the policy model via the deployment option specified for it.
Collaboration	An optional set of features is available for collaborative policy modeling and versioning, allowing fine-grained version control for Policy Modeling project development, and supporting multi-developer projects. A project repository in Intelligent Advisor Hub allows easy management and retrieval of any previous project revisions. To enable these features contact your Oracle sales representative.
Decision service references	A decision service reference is defined in Policy Modeling and is used to link a decision service in Intelligent Advisor Hub to a Policy Modeling project. Authoring a reference involves mapping Policy Modeling attributes to the decision service inputs and outputs. This allows data and logic to be defined and maintained centrally in Intelligent Advisor Hub and shared across projects.
Reference tags	A reference tag is a name that can be added to any rule or rule element to identify the source material on which it is based.
Temporal reasoning	Temporal reasoning is the ability to reason with policy model attributes or outcomes for which the values change over time. Rules written in Policy Modeling are thus time-aware, operating simultaneously both at a specific point in time, as well as across time periods.
Translations	Translation documents are Excel files in which translations are maintained of all the text that can appear when projects are deployed. This allows you to deploy a single policy model in multiple languages, while continuing to develop the policy model in its original language.
Value lists	A value list, otherwise known as an enumeration, is a tool for managing lists of potential values for an attribute in your policy model.
Forms	A form is a document that can be generated from an interview to provide the user with a record of the interview, including answers and conclusions. Forms have many uses, including pre-filled claim forms and advice letters.
Inclusions	An inclusion is a policy modeling project that has been imported into another policy modeling project for the purpose of reusing some or all of the project’s assets.
Audit reports	An audit report is a complete list of reasons about how the value of the goal attribute was determined. It is available in connected interviews without requiring any report configuration. It is most suitable for displaying to auditors.
Transcripts	A transcript saves all the interview questions asked and answers given during the interview. You can obtain a reviewable transcript for an interview, without having to use any data mapping.