There are currently three types of data transformations that you can create through the Edlink dashboard: Basic Transformations, AI Transformations, and Custom Transformations.
Basic Transformations
Basic transformations are designed to be used by non-technical members of your team. They are pre-written (by the Edlink team) transformations that can accomplish common data transformation tasks. These transformations are easy to use and require no coding experience.
Here are some examples of basic transformations:
- Infer Roles: This transformation can be used to infer the role (e.g. teacher or student) of a person based on their class enrollments.
- Hide Classes in Inactive Sessions: This transformation can be used to remove classes from the dataset that are in inactive sessions (e.g. last semester).
- Remove Inactive Users: This transformation can be used to remove people from the dataset who are not enrolled in at least one active class.
AI Transformations
AI transformations are designed to be used by semi-technical members of your team. While they do not require any coding experience, they do require some understanding of how to write a prompt that clearly articulates the desired transformation. AI transformations use OpenAI's language models to generate code that accomplishes the desired transformation.
The transformed data can be previewed in real time so that you can see the results before saving the transformation.
Custom Transformations
Custom transformations are designed to be used by technical members of your team. They require writing code in Javascript or Typescript to accomplish the desired transformation. Custom transformations are the most flexible and powerful type of transformation, as they allow you to write any code you want to transform your data.
Code is executed in a secure environment and can add, updated, or delete entities in the dataset. Custom transformations can be used to accomplish complex data transformations that are impossible (or impractical) with basic or AI transformations.