Preference Learning
This capsule discusses how to set up your capsule to use preference learning.
Preference learning is done offline and is not currently demonstrated in the Simulator. This capsule primarily shows what full modeling for preference learning might look like.
There are two primary capabilities modeled in this capsule: a search functionality for flowers (FindFlowers) and a search functionality for flights (FindTripOptions).
Because you cannot submit a capsule with the example namespace, in order to test a sample capsule on a device, you must change the id in the capsule.bxb file from example to your organization's namespace before making a private submission.
For example, if your namespace is acme, change example.preferenceLearning to acme.preferenceLearning.
FindFlowers
This functionality demonstrates how to use bootstrapped preferences, which are primarily set up through the ProductType.preference.bxb file. When returning FlowerProduct structures, it lists Bouquet and FlowerArrangement types first for users to choose from, as users typically look for these arrangements. Additionally, as the user continues to make choices when searching for flowers, eventually Bixby will learn the productTypes and flower types (through the name property) that the user prefers.
FindFlightOptions
This functionality demonstrates how to use observation-alias. When a user selects an airline's logo from the given results, the Carrier.code property is marked as a preferred option by the user. While the logo cannot be a preferable property of the Trip because it is an image, it can be associated with a property that can be considered preferable.