We want to identify cohorts of users based on their interactions history usage and affinity with polls on the Hunch App. Further, we want to send a weekly digest of list of polls to each cohort based on cohort interactions history usage and affinity.
On short term basis, we can pick a static recommendation poll list which can be sent as part of weekly digest. This can be done via following:
Pick top-5 poll creators from our dataset. Pick top-5 hunched polls for each of these creators. We get 25 list of polls which can be sent as weekly digest.
On a long term basis, we can use AWS Personalize recipes which will find segmentation of users and polls which should be part of each user segment to recommend:
AWS Personalize Solution
Following is the proposed flow of solution we would be using:
User Segmentation Model Recipes
Amazon Personalize offers two recipes that segment your users based on their interest in different product categories, brands and more.
The item affinity recipe aws-item-affinity identifies users based on their interest in the individual items in your catalog such as a movie, song, or product. The item attribute affinity recipe aws-item-attribute identifies users based on the attributes of items in your catalog such as genre or brand. This allows you to better engage users with your marketing campaigns and improve retention through targeted messaging.