Amazon web services (AWS) has recently announced a promotion feature in personalize to explicitly recommend specific items based on business rules.
Amazon Personalize enables businesses to improve customer engagement and monetization metrics by recommending personalized items to the customers. By using AWS personalize engineers could start developing recommendation engines without having prior experience with machine learning and recommendation systems.
Businesses are using promotions as one of the main strategies to increase user engagement and revenue. Promotions are considered as part of the regular recommendation to fulfill the business goals. As it is stated in the blog post:
You can use promotions to have a percentage of your recommendations be of a particular type for any application regardless of the domain. For example, in e-commerce applications, you can use this feature to have 20% of recommended items be those marked as on sale, or from a certain brand, or category. For video-on-demand use cases, you can use this feature to fill 40% of a carousel with newly launched shows and movies that you want to highlight, or to promote live content. You can use promotions in domain dataset groups and custom dataset groups (User-Personalization and Similar-Items recipes).
The following diagram shows how promotions are used in recommendation pipelines:
Recommendation with promotions
As shown in the figure, the promotion items are defined in the system catalog and then it will be loaded into the input data set. The personalized try to recommend the promotion based on the assigned percentage in the whole recommendation. As an example, once we assign the number to be 50%, then around 50% of the recommended items will come from promotion items.
In the blog post, the process of creating a promotion feature is simply described as:
Amazon Personalize makes configuring promotions simple: first, create a filter that selects the items you want promoted. You can use the Amazon Personalize console or API to create a filter with your logic using the Amazon Personalize DSL (domain-specific language). It only takes a few minutes. Then, when requesting recommendations, specify the promotion by specifying the filter, the percentage of the recommendations that should match that filter, and, if required, the dynamic filter parameters. The promoted items are randomly distributed in the recommendations, but any existing recommendations aren’t removed.
The sample promo application with the detailed steps is shared with the source code on Github.
Having recommendations is very critical for many online businesses. It motivates many cloud providers like Amazon, Google and Microsoft to provide recommendations and additional features on their platforms.
