Defining a Dataset
Dataset Definition
Three components can define any dataset. This can be helpful as a planning structure as you begin to build new datasets.
-
Activities
All data is represented in the form of activities. In order to build a dataset, you first need to identify which activities have the information and customer behaviors you care about to answer your question. -
Occurrence
Do you want the first activity, the Nth (second, third, etc...), the last one, or all of them? -
Temporal Joins
Previously called "Relationships"
Temporal Joins define how you incorporate additional activities into your dataset. They enable you to enrich your dataset with information about specific customer behaviors.
![Screen Shot 2020-07-10 at 2.22.20 PM.png 1606](https://files.readme.io/3b556de-Screen_Shot_2023-08-11_at_11.03.25_AM.png)
Dataset Definition
Filtering, Aggregating, and Deriving New Columns
Once the dataset structure is defined, you can modify it further by:
- Filtering rows
- Deriving computed columns
- Aggregating the dataset
Start building datasets!
Browse the Dataset Tutorials for step-by-step examples.
Still have questions?
Our data team is here to help! Here are a couple ways to get in touch...
π¬ Chat us from within Narrator
π Email us at [email protected]
π Or schedule a 15 minute meeting with our data team
Updated 12 months ago