Streaming service provider Biscalli has developed a recommendation engine to

recommend content to its subscribers based on their content history, ratings for that

content, and the device on which that content was viewed. The history and ratings are

used in an input layer within the engine that stack$-$ranks similar content. That content

is then assigned weights within another hidden layer, and recommendations are made

to the subscriber that target certain types of content on specific devices in an output

layer. This business process most likely uses a recommendation engine based on:

Decision trees.

Logistic regression.

Neural networks.

K$-$means clustering.