Glossary
This glossary provides terms and definitions for Decision Intelligence Client Managed Software
Bold terms in definitions indicate other terms defined in the glossary.
A
- action task node
- In a ruleflow, a task node that contains rule action statements. These action statements are executed each time the task node is called.
- attribute
- A characteristic of a composite type. For example, the brand of a car is one of the car attributes. An attribute has a data type and a verbalization. It can have one or more values.
B
- branch
- An independent line of development. It contains decision services and their change history for this branch.
- business rule
- A policy, constraint, or required operation that applies a specific set of business conditions or dependencies.
C
- collaborator
- User who works on a decision automation with other users.
- composite type
- A data type that has one or more attributes.
D
- data model
- A data model is a collection of data types and their associated verbalization.
- data type
- The representation of a business concept. A data type has a name and a verbalization. The application provides predefined data types. Users can define custom data types, which can be enumeration types or composite types.
- decision artifact
- Data models, decision models, task models, generative AI models and predictive models are decision artifacts.
- decision automation
- A container for decision services. A decision automation is synchronized to a branch in a remote repository.
- Decision Designer
- The authoring environment where users can develop, run, and deploy decision services.
- decision logic
- A set of business rules and/or decision tables that is used to make a decision based on some input data.
- decision model
- The implementation of a decision. It contains a diagram and a decision logic and can be used as a function in other decision models and task models.
- decision node
- In a decision model, the graphic representation of an end decision or subdecision. This decision has a name and a data type. It contains a decision logic that uses inputs from other decision nodes or input data nodes to produce an output that is the result of the decision.
- decision operation
- An entry-point to a model. It contains a reference to a model and a function, and is used to create versions.
- decision runtime
- A server that exposes decision services through decision service endpoints.
- decision service
- A set of related decisions that are managed and deployed together. A decision service contains decision artifacts and decision operations.
- decision service archive
- A collection of compiled decision source files. A decision service archive is compiled by the build orchestrator from decision source files. This archive is then pushed to a deployment space, from where it can be loaded by the decision runtime. At this point, the archive becomes part of a runtime decision service, which is the web service that can be called from a client application.
- decision service endpoint
- The URL where the decision service can be accessed by a client application. This endpoint is exposed by the decision runtime.
- decision table
- A decision table represents several business rules that share the same logic but apply different conditions and actions. Each row in a decision table forms a complete business rule.
- deployment space
- A set of decision service archives and their associated metadata. A deployment space can be dedicated to a development phase or a group of users.
- diagram
- In a decision model, the visual representation of a decision.
E
- end decision
- The final decision that is made in a decision model.
- enumeration type
- A data type that contains a list of values.
F
- folder
- A task model artifact that can be used to organize to group and organize other artifacts, such as business rules and decision tables.
- function
- The list of input and output parameters needed to execute a decision model, a predictive model, or a specific ruleflow within a task model. It is used to create decision operations, and for model composition.
- function node
- In a decision model, the graphical representation of a function. It provides values that are computed in another decision model within the same decision service.
- function task node
- In a task model, the graphical representation of a function. It provides values that are computed in a decision model, a predictive model, or another task model within the same decision service.
G
- generative AI model
- A model that leverages large language models (LLM) to produce content that aligns with the given instructions and context.
I
- input data node
- In a decision model, the graphic representation of data that is used to make decisions. This data has a name and a data type.
- interaction policy
- Interaction policies define how business rules interact with each other by governing their order of execution. Interaction policies apply at the decision node level.
P
- predictive model
- A model that implements a prediction based on a machine learning model. A predictive model returns a probability or score that can be used in decision models.
- predictive node
- In a decision model, the graphical representation of a prediction. It provides predictions that are computed in predictive models.
R
- remote repository
- A repository hosted on a Git server that is synchronized with a project. Decision source files are pushed to the remote repository before being built and deployed.
- rule language
- A language for expressing rules with natural language terms and syntax. It uses user-defined vocabulary that expresses business concepts.
- rule task node
- In a ruleflow, a task node that refers to rule artifacts and orders them.
- ruleflow
- A method of controlling and ordering the execution of rule artifacts. A ruleflow is defined in terms of task nodes.
- run
- The process of checking that a decision model, predictive model or task model works the way that is expected. Users define test data and run the decision model with it to see the decision results.
S
- share
- Committing changes to a project. When a user shares changes, these changes become available to other users who work in the same project.
- subflow task node
- in a ruleflow, a task node that refers another ruleflow. A subflow tasks node can reference a ruleflow in the current task model, or in another task model in the same decision service.
T
- task model
- The implementation of a decision. It contains rule artifacts and a ruleflow to organize their execution.
- test data
- The data that users enter to validate that the decision logic works as expected.
V
- variable set
- A group of variables that can be used across a task model.
- verbalization
-
(v.) The process of associating terms and phrases to data types and attributes to use them in a decision.
(n.) The phrase defined by the user to refer to a data type or an attribute. Each verbalization is defined for a specific locale, so a data type or an attribute can have more than one verbalization. E.g: the data type
birthDatecan be verbalized asbirth dateanddate de naissance. - version
- A pointer to a specific point in the change history. A version has a name and a description. A version is reflected in the remote repository as a Git tag.