- 최신
- 최다 투표
- 가장 많은 댓글
Hi, For recipes that are delivering real-time personalization, Amazon Personalize immediately uses real-time interactions between a user and existing items or actions (meaning items, actions and users present in the latest training datasets) when generating recommendations for the same user. Here is the list of recipes that support real time personalization:
- Recommended for you (ECOMMERCE use case)
- Top picks for you (VIDEO_ON_DEMAND use case)
- User-Personalization recipe
- Personalized-Ranking recipe
- Next-Best-Action recipe As Trending-Now recipe calculates the increase in interactions that each item has over configurable intervals of time, it will re-calculate based on the interval rather than every time an interaction happens. For new items and new users, that are not present in your latest data you can read more info here: https://docs.aws.amazon.com/personalize/latest/dg/how-new-data-influences-recommendations.html . Best, Nensi
Adding to Nensi's response: The recipe that will update recommendations in real-time by passing item interaction data using PutEvents is user-personalization.
The user-personalization recipe provides real-time personalization, which means it can immediately use new item interaction data passed through PutEvents to generate updated recommendations for a user.
https://docs.aws.amazon.com/personalize/latest/dg/event-record-api.html https://aws.amazon.com/blogs/machine-learning/implement-real-time-personalized-recommendations-using-amazon-personalize/
