Layered Reasoning with Dependent Variables
Brim enables layered reasoning through a concept called Dependent Variables.
Dependent variables can make decisions based on the values of multiple variables or dependent variables.
Fields
Name.
This name should be descriptive.
Variable type.
This describes the type of data the system should return. The options are:
- Text. No restrictions on format.
- Boolean. The value must be True or False.
- Integer. The value must be a whole integer or number.
- Float. The value is a number that could be whole or fractional.
Instructions.
This is the most important part of the dependent variable definition. It tells the AI what it means to correctly label the variable. A good instruction includes:
- A clear definition of the variable.
- Semantics necessary to guide how to abstract values
- Temporal considerations. Is this historical? Current?
- One or more examples.
Variables/Dependent Variables
- A list of variables and dependent variables that the dependent variable should use as input to its decision.
Default value for empty response.
Dependent variables return one value per patient. The default value for empty response determines what the LLM should respond for a patience when it finds no relevant evidence. This defaults to "None", but could be "False", "No evidence", or another value depending on your application.
Optional/Advanced Fields
Option definitions.
If your variable has a limited set of options (for example, "Stage I", "Stage II", and "Stage III"), define them here. Don't forget to include options for unknown or not found.
Prompt definition
This allows you to edit the surrounding prompt provided to the LLM. Usually the default prompt is fine.
More about Dependent Variables
You can create dependent variables directly in Brim Analytics or import them via a CSV.
Brim can automatically create a dependent variable and the required input variables. This is handy for setting up complex logic like clinical trials.