Today, Amazon Personalize announces the general availability of two new recipes, User-Personalization-v2 and Personalized-Ranking-v2 (v2 recipes). Built on Transformers architecture, these new recipes support catalogs with up to 5 million items with lower inference latency. Amazon Personalize testing showed that v2 recipes improved recommendation accuracy by up to 9% and recommendation coverage by up to 1.8x compared to previous versions. A higher coverage means Amazon Personalize recommends more of your catalog. These new recipes also support item metadata like genres and descriptions in inference responses, allowing customers to easily enrich recommendations in their user interfaces.
Source:: Amazon AWS