What is a Personalize "context-aware solution" ?

Question

[This blog post][1] about "contextual recommendations" says this feature requires a "context aware solution  with a User Personalization or Personalized Ranking recipe".  In the blog, hrnn-metadata is not listed as a compatible recipe yet it is used it in the demo afterwards.

**What is a context aware solution? Which Personalize recipe can learn from interaction context?**

[1]: https://aws.amazon.com/blogs/machine-learning/increasing-the-relevance-of-your-amazon-personalize-recommendations-by-leveraging-contextual-information/

Accepted Answer

Context - is metadata in an interaction dataset (up to five metadata fields). 
You could use any metadata based recipe (HRNN-metadata(legacy), HRNN-cold(legacy), Personalized-Ranking, User-Personalisation).

Context aware solution is when you could pass context (device, location) as an attribute when you ask for a recommendation and Personalize will provide recommendation with attention to a context (For example, Amazon Personalize can provide different recommendations when a user is browsing on a mobile device than when that same user is browsing on a desktop.)

https://docs.aws.amazon.com/personalize/latest/dg/how-it-works-dataset-schema.html

What is a Personalize "context-aware solution" ?

相關內容