Datasets and Schemas

There are three types of datasets.

  • Users - This dataset stores user keys and metadata. Metadata can include data such as gender, age, or income bracket.
  • Items - This dataset stores item keys and metadata. Metadata can include data such as price, colour, or SKU.
  • Interactions - This stores the event data between your users and items.

Including metadata improves the accuracy of recommendations and is especially helpful for enhancing cold starts.

An application can have only one of each type of dataset. Before you create a dataset, you define the schema for that dataset. A schema tells Xineoh Consumer Behaviour API about the structure of your data and allows it to process the data correctly. A schema has a name key whose value must match the dataset type.

For an application, each dataset type has required-fields in the schema. Once you have defined the schema and created the dataset, you cannot make changes to the schema.