Skip to content

Latest commit

 

History

History
331 lines (261 loc) · 47.9 KB

PersonalizeCheatSheet2.0.md

File metadata and controls

331 lines (261 loc) · 47.9 KB
Raw

Amazon Personalize Cheat Sheet

Personalize Terminology

  1. Dataset Groups: This refers to a group of five datasets, only one of which (item-interactions) is required. They are as follows;

a) Item-interaction data (consisting of at minimum a timestamp, a user_id, and an item_id) represents positive interactions between users and items at a specific time. In addition to this dataset you can optionally bring four additional datasets. Items metadata, user metadata. Personalize assumes that all interactions are positive so if you have examples of user interactions with items that are not positive (such as customers leaving a bad review of an item) you should not include those interactions in your dataset. This is the only required dataset for most use cases.

b) Items metadata (required for similar items) can consist of one textual field (note if the text field is not complete it is recommended to fill in any null fields with a placeholder such as “blank”) and up to 100 item metadata fields. Consists at minimum of an item_id field and one metadata field. creation_timestamp controls the eligibility of new items for exploration (in which they receive a bonus to their score which is inversely proportional to the number of interactions/impressions the item has in the dataset)

c) User metadata Not required for any use cases but can optionally be added. Can consist of information about users that rarely changes for information about users that rapidly changes (such as the users location or what device type they are logging in on) consider using contextual metadata instead.

d) Action metadata Required only for the Next Best Action recipe, not used in other recipes, consists of at minimum an action_id field. Can consist of additional fields such as creation_timestamp which controls the eligibility of new items for exploration, repeat_frequency, which is the number of days after recommending an action to wait before recommending it again and action_exploration_timestamp which represents the time (in unix epoch time) after which to cease recommending an action.

e) Action-interaction data Optional but recommended for Next Best Action use case consists at minimum of a timestamp, a user_id, an action_id and an event_type representing a user taking an action (or choosing not to take an action when prompted) at a specific point in time. The event_type contains information on the users response and can consist of “taken”, “not taken“, and “viewed“.

  1. Recipe: This refers to a specific type of model structure designed for a specific family of business use cases. For example a user personalization recipe will produce item recommendations that are likely to appeal to an individual user.

  2. Solution: A combination of dataset group and recipe - basically the framework to train a model for a specific use case.

  3. Domain Recommenders vs Custom A domain recommender is a recommender optimized for a use case in one of two specific domains: E-commerce or Video on Demand to facilitate the easier onboarding in these cases. A custom solution supports a broader array of use cases but leaves a-lot of configuration up to the developer.

  4. Solution Version: A solution for which a contained model has been trained for a specific use case, can be used to run batch recommendation (inference) jobs or deployed as a campaign.

  5. Campaign: A deployed solution version consisting of an api with a provisioned transaction capacity for generating real-time recommendations. After the campaign finishes deploying you can get recommendations with the GetRecommendations or GetPersonalizedRanking operations.

  6. Event Tracker: Can be connected to an existing dataset group with an interactions dataset. Allows users to feed additional events to the interactions dataset in near real time (usually in under 2 seconds). If you have deployed User-Personalization, or Personalized-Recipes or Recommended for you (E-Commerce), or Top picks for you (Video on Demand).

  7. Domain Recommenders: When used in the context of Amazon Personalize this refers to a bunch of preconfigured solutions that streamline the e2e process. Unlike in custom solutions where developers need to go through several steps to deploy an API capable of serving real time predictions Domain Recommenders create and deploy model serving API’s directly from a dataset group they also handle the deployment of some relevant filters and some aspects of maintenance of a deployed model.

  8. Contextual Metadata: This refers to data accompanying an interaction that we know ahead of time such as the type of browser or device a client is logging in during or if the interaction is occurring during any sales or marketing events. When training a solution version / recommender. To be incorporated into a model this data needs to both be included in our historical interactions and fed to personalize in real time through both an event tracker and in the GetRecommendations request.

Is Amazon Personalize a Good Fit?

Amazon Personalize is a great platform for operating a recommendation system at scale on AWS but it is not appropriate for any and all personalization or recommendation scenarios. The chart below is a rough guide for good and bad fits. Generating “Next Best Offer” is not supported by the service.

Good Fit Bad Fit
Recommending items to known users. Movies to users based on their watching history or items to retail users. Personalize will update to new items being inserted into its dataset within 2 hours. Recommendations based only on explicit user or item metadata flags and their is no record of user history
Recommending items to new users based on their in session behavior. Low data volumes for users, items, and interactions (see table below).
Recommending new items to new users. Personalize can integrate new items into its reccomendations within hours Personalize does not support explainability (exposing why a user is getting certain recommendations and customers cannot bring their own models. For these two features, sagemaker is a better approach.
Media, Retail, Publishing and Travel & Hospitality are a good fit. Finance, Insurance companies with limited inventory (types of insurance, bonds etc)
Companies whose technical teams are understaffed with large project backlogs - Personalize is MUCH quicker to deploy than a custom built recommendation system Use cases that where you will only be recommending a small percentage items in your interaction history to your users (such a financial service use case where you have a history of bank transactions and account openings but are only interested in a recommendation on what sort of new account to recommend to a user).
Companies who already track clickstream events - this makes using Amazon Personalize and benefiting from its ability to update user reccomendations in near real time easier Companies who currently don't track clickstream events and lack the technical capacity to feed interactions to Amazon Personalize as they occur
Companies with large item catalogs (videos, products, content whatever) where the typical user journey involves a search through the catalog of a reasonable length before an item is purchased / viewed Companies who are soley intrested in recommending very new items with short shelf lives to their customers. Personalize works well in publishing and news for example but can't replace editorial curation of a breaking news section without substanital effort
More than 50 users and 50 items and 1000 interactions with items that are still active in your dataset More than 5M active items and you need to be capable of recommending them all to a customer

Use Cases By Recipe

What kind of use cases can be solved and how?

  1. Personalized Recommendations User-Personalization/Recommended For You/Top Picks for You:
    1. This is a primary use case of Amazon Personalize, using user-item-interaction data to build a recommendation model that directly targets each user, and allows for new users to be added on the fly with PutEvents without re-training. PutEvents also allows for the users to see recommendations that are based on their most recent behavior so you do not lose out on that additional info. Also you can feed in context specific components like device type or location to improve results.
    2. You can also add item and user metadata in order to better enrich the model, or to filter recommendations by attribute.
    3. For Video On Demand and Retail use-cases, the domain recommenders “Top picks for you” and “Recommended for you” allow you to get up and going quickly and with less operational overhead.
  2. Recommending Items to New Users User-Personalization/Recommended For You/Top Picks for You:
    1. New users (aka cold users) can be added to your existing User-Personalization solutions by leveraging the PutEvents feature. Each new user starts with a representation in the service that returns the popular items. This representation is shifted by the user’s behavior. As they interact with content within the application and the events are sent by the application to Personalize, recommendations are updated without having to re-train the model. This delivers up to date personalization without constant retraining.
    2. Metadata for new users can be added to the Users dataset using the PutUsers API or dataset import jobs. This data is incorporated into the model at the next retraining.
    3. Contextual metadata for new users, such as their current location or device type, can be provided when retrieving recommendations to influence recommendations.
  3. Recommending New Items User-Personalization/Recommended For You/Top Picks for You:
    1. This is incredibly useful when you have new items (aka cold items) that need to be showcased to users with some form of personalization. This allows for the items to be recommended without historical precedent based on metadata factors.
    2. This can also be used with incrementally updating your Items dataset to more easily cold-start new items.
    3. Lastly, this approach leverages an exploration capacity to help you quickly determine which results make sense, and which do not for recommendations, a far better approach than just blindly pushing new content.
  4. Re-Ordering by Relevance Personalized-Ranking:
    1. Uses the same HRNN algorithm underneath User-Personalization but takes in a user AND a list of items. This will then look at the collection of items and rank them in order of most relevant to least for the user. This is great for promoting a pre-selected collection of items and knowing what is the right thing to promote for a particular user or for personalizing search results.
  5. Related Items Similar-Items/SIMS/Because you watched X/More like X/Frequently bought together/Customers who viewed X also viewed:
    1. Similar-Items: Deep learning model that considers both interaction data and item metadata (optional) to balance related item recommendations based on interaction history and item metadata similarity. Useful when you have less interaction data but have quality item metadata or when you frequently introduce cold/new items.
    2. SIMS: Pretty simple idea, implemented via item-item collaborative filtering but basically looks at how people are co-interacting with particular items and then determine how similar items are at a global level based on interaction data. Does not consider item or user metadata and is not personalized to each user. Useful when you have a lot of relevant interaction data, do not have a lot of cold items (changing catalog), and/or lack item metadata.
    3. For Video On Demand and Retail use-cases, the domain recommenders “Because you watched X”, “More like X”, “Frequently bought together”, and “Customers who viewed X also viewed” allow you to get up and going quickly and with less operational overhead.
  6. Frequently Bought Together Similar-Items/SIMS/Frequently bought together:
    1. The key is preparing the right data used to train a model in Personalize and choosing the right recipe. For example, train a SIMS model on purchase data only and, if possible, train only on purchase data where customers have purchased multiple items and/or purchased items across multiple categories. This will bring the desired behavior to the model and create diversity in recommendations (which is what you want for this use case).
    2. SIMS can also be combined with Personalized-Ranking to rerank recommendations from SIMS before presenting them to the user. This provides a personalized order of items that are frequently bought together.
    3. The retail domain recommender “Frequently bought together” will allow you to get up and going quickly and with less operational overhead.
  7. Most Popular Overall Popularity-Count/Most-Popular/Most-Viewed/Best-Sellers:
    1. Not machine learning, just a baseline from counting the most commonly interacted with items. This recipe is useful for popular item recommendations or for creating a baseline of offline metrics that can be used to compare against solution versions created using other user-personalization recipes with the same datasets.
    2. For Video On Demand and Retail use-cases, the domain recommenders “Most popular”, “Most viewed”, and “Best sellers” allow you to get up and going quickly and with less operational overhead.
  8. Trending Items aws-trending-now:
    1. Also not machine learning - represents items for which the difference in the number of interactions occurring in the most recent 30 minute time-step and those occurring in the previous 30 minute time-step is the greatest. This is designed to capture items that are starting to go viral not those that are the most popular in the catalog (for that use one of the Most Popular Overall recipes)
  9. User Segmentation Item-Affinity/Item-Attribute-Affinity:
    1. Create segments of users based on their affinity to specific items in your catalog or their affinity to item attributes. Excellent match with marketing campaigns where you’re looking to target users who will have an interest in specific items you’re looking to promote or items similar to existing items.
  10. Next Best Action aws-next-best-action
    1. Create a model which predicts the likelihood a user taking a particular action at a point in time. Can be setup to predict the likelihood of a user taking an action by themselves or when prompted (recommended). Unlike other personalize recipes the score in this case is intended to be a human understandable probability suitable for downstream use in tasks like expected value calculations.

Killer Features:

  1. Domain Optimized Reccomenders: recommenders for Video On Demand and Retail use cases
    1. A Domain dataset group is an Amazon Personalize container for domain specific pre-configured resources, including datasets, recommenders, and filters. Use a Domain dataset group if you have a streaming video or e-commerce application and want to let Amazon Personalize find the best configurations for your recommenders.
  2. Contextual Recommendations
    1. Allows you to fine-tune recommendations based on state that varies with the interaction rather than specific to the user or item. Think the user’s current location, device/channel being used, time of day, day of week, etc.
    2. See this blog post for a detailed example: https://aws.amazon.com/blogs/machine-learning/increasing-the-relevance-of-your-amazon-personalize-recommendations-by-leveraging-contextual-information/
    3. See this blog post for an example of how to use CloudFront to collect this sort of context to populate Get Recommendation requests in near real time: https://aws.amazon.com/blogs/machine-learning/recommend-and-dynamically-filter-items-based-on-user-context-in-amazon-personalize/
  3. Interaction and Metadata Filtering
    1. Filter recommendations based on the user’s interaction history or the metadata attributes for the items or current user. Very handy in nearly all Media or Retail workloads. For example, exclude recently purchased or out of stock items or include/exclude recommended items based on category or genre.
    2. See this blog post for details: https://aws.amazon.com/blogs/machine-learning/enhancing-recommendation-filters-by-filtering-on-item-metadata-with-amazon-personalize/
  4. Promotional Filters
    1. Promote specific items in recommendation responses based on a filter expression and control how many items in the recommendation response are promoted. For example, require that 30% of the recommended items have been recently released/published (where you control the timeframe) or require that 25% of the recommended items are private label or currently on sale.
    2. See this blog post for details: https://aws.amazon.com/blogs/machine-learning/customize-your-recommendations-by-promoting-specific-items-using-business-rules-with-amazon-personalize/
  5. Batch Inference
    1. Great for exporting large quantities of recommendations to files for caches, for email campaigns, or just general exploration.
  6. AutoScaling Campaigns/Recommenders
    1. For real time inference endpoints (campaigns and recommenders), the service will automatically scale to meet your traffic demands if a particular campaign or recommender is over-subscribed. It will then also reduce to your requested minimum capacity when the traffic volume abates.
  7. Unstructured Text as Item Metadata
    1. Add your product descriptions, video plot synopsis, or article content as an item metadata field and let Personalize use natural language processing (NLP) to extract hidden features from your text to improve the relevance of recommendations.
  8. Put Events
    1. Allows for applications to update Personalize in real time with changes in intent from user behavior. This means that each subsequent request can adapt to that intent WITHOUT retraining.
  9. Put Items/Put Users
    1. Allows for applications to add/update individual or mini-batches of items or users without having to upload the entire items and users datasets.
    2. See FAQs below for more detail.
  10. KMS Integration
    1. All data can be encrypted using a customer managed key, all data is encrypted regardless.
  11. No Information Sharing
    1. All customer data is fully isolated and is not leveraged in order to improve the recommendations of Amazon or any other party.
    2. Models are private to the customer’s AWS account.

Video Series:

  1. Introduction To Amazon Personalize: https://youtu.be/3gJmhoLaLIo
  2. Understanding Your Data with Amazon Personalize: https://www.youtube.com/watch?v=TEioktJD1GE
  3. Solving Real World Use Cases with Amazon Personalize: https://www.youtube.com/watch?v=9N7s_dVVWBE
  4. Getting Your Amazon Personalize Recommendations in Front of Your Users: https://www.youtube.com/watch?v=oeVYCOFNFMI
  5. Get Your Amazon Personalize POC to Production: https://www.youtube.com/watch?v=3YawVCO6H14
  6. Enhancing Customers Experiences with Amazon Personalize (including a demo of the Retail Demo Store): https://youtu.be/71dWFhzw7iw
  7. Personalized Search: https://www.youtube.com/watch?v=w8Ut02535_I&ab_channel=AmazonWebServices

FAQs:

  1. How often should I retrain?
    1. Retraining frequency is determined by the business requirements. How often do you need to learn globally about your users and their behavior with items? How often do you need to include new items? The answers determine how frequently you should train. Generally speaking, most customers retrain weekly. See below for more detailed guidance.
    2. If using the “aws-user-personalization”, “recommended-for-you”, or “top-picks” recipes, the service will automatically update the solution version in the background every 2 hours (at no additional cost). This auto update process will bring in new items added since the last update so they can start being recommended to users (i.e. cold starting items). This works in coordination with the explorationWeight parameter set on the campaign/recommender to control the weight placed on recommending new/cold items vs relevant items (explore/exploit).
    3. If the 2 hour auto-update is not frequent enough for introducing new items, you can manually create a new solution version with trainingMode=UPDATE and update the campaign more frequently (i.e. hourly). This essentially does the same thing as the auto-update, just on a custom frequency. However, there is a cost for training hours for doing this manually.
    4. Regardless of the auto or manual update mode process taken, this does not fully retrain the model. It is still necessary to occasionally create a new solution version with trainingMode=FULL to fully retrain the model. This is important to do occasionally to recalculate the weights across the model based on all data but the auto-update process makes full retraining necessary less frequently. This is where the weekly guidance comes in. So let auto update run all week and do full retraining once a week.
    5. To be more precise in retraining frequency, another approach is to monitor online metrics. When they start to trail off (i.e. model drift), it may be time to retrain.
  2. How do I add a new user?
    1. If you are using the PutEvents API, the new user exists as soon as you log their first action. If you are not leveraging this, then the user will exist in the system as soon as you have retrained a model that contains their behavior in your interactions dataset.
    2. If your user is not known (a new anonymous user before signup) you can still work to cold-start them. If you can assign a new UUID for their user and sessionID immediately, then you can continue the process as defined above to cold start a user.
    3. If that path does not work, you can still generate a new UUID for the sessionID, call PutEvents without a userID, and then continue to specify the same sessionID after a valid userID has been generated for them. When you retrain, Personalize will combine the historical data, with PutEvents data, and when it sees matching sessionIDs it will combine all prior anonymous interactions together with non-anonymous interactions for the user. This would allow you to specify the history before they had a valid internal userID.
    4. You can add/update users individually or in mini-batches with the PutUsers API. However, only users with interactions will receive personalized recommendations either after (re)training or when cold started with the PutEvents API.
  3. How do I add a new item?
    1. There are two ways to add items to the items dataset: 1/ Add new items to the items dataset by uploading the full dataset using a dataset import job, or 2/ add items individually or in mini-batches using the PutItems API.
    2. New items will be incorporated into recommendations after retraining if interactions also exist (all recipes) or cold-starting new item recommendations with or without interactions after the solution has been updated (trainingMode = FULL/UPDATE for aws-user-personalization and HRNN-Coldstart only).
    3. For example, you can organically stream new items into your historical dataset by placing banners for new releases. Anything that causes a user to interact with the new items and for that action to be logged can improve recommendations after the next training.
  4. How do I filter results for certain conditions?
    1. Using the filters feature for either Interaction (https://aws.amazon.com/blogs/machine-learning/introducing-recommendation-filters-in-amazon-personalize/) or Metadata information ( https://aws.amazon.com/blogs/machine-learning/enhancing-recommendation-filters-by-filtering-on-item-metadata-with-amazon-personalize/ ).
    2. Filtering based on interaction history currently only considers the most recent 100 real-time interactions (PutEvents API) and the most recent 200 historical interactions in the dataset at the time of retraining. All event types are included in the 100/200 limits. This can be raised to 2000 per event type by filing a service ticket
  5. I need to filter items based on a rolling date value but filters do not support dynamic values for range operators. What are my options?
    1. Filters now support dynamic values with range operators (i.e. <, <=, >, >=). Add date values expressed as Unix timestamps to your Items dataset or reuse the CREATION_TIMESTAMP reserved column for filtering on item age. Then create a dynamic filter that compares a passed in date value (passed in at inference using the filterValues parameter) to the date value for items. For example, INCLUDE ItemID WHERE Items.CREATION_TIMESTAMP > $TIMESTAMP where $TIMESTAMP can be a date from the recent past.
  6. Why should I use Amazon Personalize over a custom solution?
    1. Assuming your data and use cases are aligned, this is a great way to get a best in class model in front of the end users faster. While handling the operational burden of running a recommendation system at scale, Personalize also frees you up to improve feature engineering, data collection, user experiences, or solve other problems.
  7. I have a use case where my customers purchase or interact with items in my catalog infrequently (e.g. purchasing a car). Is Personalize still a good fit?
    1. Yes, Personalize can still be put to use effectively for this type of use case. For example, a Similar-Items model can be trained based on all user activity such as browsing or purchase history (online and/or offline) and then used for related item recommendations on item detail pages. This allows you to leverage recent activity across all active users to make relevant recommendations to anonymous/returning/cold users.
    2. Real-time recommendations are also effective here since Personalize is able to learn from a user’s current interest and adapt recommendations quickly. For example, recommend popular items initially and then quickly personalize recommendations after a few interactions are streamed using the PutEvents API.
  8. Should I use AutoML?
    1. No, the recipes solve different use cases. Take the time to select the most appropriate recipe for the use-case and skip this feature.
  9. Should I use HPO / How often?
    1. Infrequently. Take the results of one HPO job and use them explicitly in the solution config for several retrainings. Then run HPO again and repeat. Realistically tuned parameters should not shift much between training jobs. This approach will keep your training times, and therefore costs, lower than running HPO for all training jobs without sacrificing model quality.
  10. How can I predict the pricing of training?
    1. Unfortunately there really is not a great way to know this beforehand, we do have some testing that has been done on the MovieLens dataset. For example using User-Personalization it takes around 6 human hours for the training on 25M interactions but less than 1 human hour to train on 100k interactions. Because training is sharded over multiple hosts, the actual hours are, 53.9 hours for 50M and 2.135 hours for 100k. Billing is done on the actual hours, not the human ones.
  11. What is a TPS hour and how does it relate to pricing / usability?
    1. Amazon Personalize spins up dedicated compute resources that will remain provisioned in order to meet your expected minimum throughput requirements (Transactions per Second or TPS), they are billed in terms of hours that these resources are allocated, thus a TPS-Hour. 1 TPS-Hour is the amount of compute capacity required to deliver 1 recommendation per second for an entire hour.
    2. Usage is measured in 5 minute increments where the maximum of the average number of requests and the minimum provisioned throughput in each increment is used as the TPS-Hour value. Therefore, when the service scales above the minimum provisioned TPS, the customer is only billed for capacity actually consumed. The TPS-Hours for all 5 minute increments are summed during the billing period to determine the total TPS-Hours for billing calculations.
    3. The service will automatically scale up for you if your traffic exceeds the minimum provisioned TPS on the campaign, and it has proven to be a valuable tool for many of our customers. A capacity buffer is allocated above the minimum provisioned TPS to allow the service to absorb increases in request load while it scales out.
    4. If your customer knows they are going to have a spike of activity such as a flash sale or promotional event, have them use some automated process to update the provisioned capacity to meet the new need, then throttle it down later if they cannot wait 5-10 minutes for the service to auto-scale for them.
    5. The Amazon Personalize Monitor project provides a CloudWatch dashboard, custom metrics, utilization alarms, and cost optimization functions for Personalize campaigns: https://github.com/aws-samples/amazon-personalize-monitor
  12. How can I tell if a Personalize model is providing high quality recommendations?
    1. Personalize provides offline metrics for each solution version that measure the accuracy of predictions from the model against held out data from the interactions dataset. Use these metrics to provide a directional sense of the quality of a solution version against other versions.
    2. Online testing (i.e. A/B testing) is always going to be the best measure of the impact of a model on business metrics. If you do not have an A/B testing tool already in place, consider using Amazon CloudWatch Evidently.
    3. When you are comparing Personalize models against an existing recommendation system, all the historical data is initially biased towards the existing approach. Often the offline metrics do not reflect what a user MAY have done if they were exposed to something else (how could they, the data does not reflect it). So worth noting this effect, and the bandit based exploration Personalize can do to organically learn better from your users. Therefore running an online test for a few weeks before actually starting a test to measure results is recommended.
    4. See this blog post for details: https://aws.amazon.com/blogs/machine-learning/using-a-b-testing-to-measure-the-efficacy-of-recommendations-generated-by-amazon-personalize/
  13. How can I optimize for cost?
    1. DO NOT USE AUTOML!
    2. DO NOT START WITH HPO - build something working first, optimize last.
    3. Re-Train based on business requirements only. See FAQ question for details.
    4. Rely heavily on auto-scaling by setting the minimum provisioned TPS low unless it negatively impacts your throughput / latency targets.
    5. Consider using batch recommendations when the use-case aligns with a downstream batch process such as email marketing. Since batch recommendations run against a solution version, they do not require a campaign.
    6. The Amazon Personalize Monitor project provides some cost optimization features for optimizing campaign provisioning as well as alerting and deleting idle/abandoned campaigns: https://github.com/aws-samples/amazon-personalize-monitor
    7. You do not need to draw a new series of recommendations every time a page is refreshed. Use clever caching strategies. Users will likely find it unpleasant if a large number of assets on a page change merely because users view details on an individual product or piece of content. In addition new users will have very similar recommendations the first time they visit a page. Recommendations for each device type and other piece of contextual metadata and their unique combinations can be cached and shown to a new user at the start of a session. A personalized recommendation can then be drawn after some interactions are collected for the user.
  14. How do I provide high quality recommendations for new (cold-start) users and items in the user-personalization recipe?
    1. For cold-start users higher quality recommendations can be generated by feeding contextual metadata in with the get recommendation request (what device type the user is logging in from, where they are logging in from ect). Personalize also updates to new user interactions fed to it through putEvents in near real time so while the initial recommendation quality for a new user may not be ideal - much higher quality recommendations can be shown to the user during the session.
    2. For cold-start items submit them to personalize through the put items function. These items will be picked up by personalize within ~2hours worst case - usually faster. If exploration is enabled new items will receive a bonus to their score so that they are shown to more users.
  15. How does Exploration work?
    1. Exploration works as follows:
      1. Items are separated into items eligible for exploration and items ineligible for exploration. This eligibility is determined by comparing the items creation_timestamp (in the items metadata) to the current time minus the explorationItemAgeCutoff (a parameter set for the campaign) in days. If the creation timestamp is within that window the item is eligible for exploration.
      2. The item receives a bonus to its estimated user affinity score which is inversely proportional to its interactions count / impressions count. This count is updated approximately every 2 hours, new items are also picked up at this time
      3. The item also has a small amount of noise added to it to represent potential uncertainty in the user-item affinity learned so far and to assist with discoverability
  16. How do I use contextual metadata?
    1. Contextual metadata is additional information known about the user, or the session at the time a recommendation request is made that might change a users preferences. Examples include the weather or location for food delivery, ongoing sales and other events for retail use cases and the device type a user is logging in on for media use cases.
    2. Contextual metadata must be defined in the schema and supplied as part of the historical interaction data that the model is trained on. It MUST also be supplied as part of a getRecommendations API request.
    3. Contextual metadata is not for information about the interaction that is only known after the interaction such as what kind of discount the item was purchased at or the quantity of an item purchased.
  17. What are the best ways to use caching with Amazon Personalize? How should I integrate Personalize with my existing applications?
    1. Check out the Personalization APIs solution: Real-time low latency API framework that sits between your applications and recommender systems such as Amazon Personalize. Provides best practice implementations of response caching, API gateway configurations, A/B testing with Amazon CloudWatch Evidently, inference-time item metadata, automatic contextual recommendations, and more.
  18. What is the best way to compare Personalize to an existing user experience or another recommendation system?
    1. A/B testing is the most common technique for evaluating the effectiveness of Personalize against online metrics. Amazon CloudWatch Evidently is an A/B testing tool from AWS that can be used with Personalize. The Personalization APIs project provides a deployable solution and reference architecture.
  19. How do incremental records influence recommendations for the current user?
    1. Amazon Personalize allows you to import interactions, users, and items incrementally. These can affect recommendations for the current user in different ways depending on whether the a new solution version has been trained and what type of trainingMode was used see https://docs.aws.amazon.com/personalize/latest/dg/how-new-data-influences-recommendations.html
Increment Recipe No retraining Retraining trainingMode=UPDATE Retraining trainingMode=FULL Comments
putEvent with new user User personalization Personalization begins after 1 event, but will be more visible after ~2-5 events with 1-2 seconds delay after the PutEvents call after each event is recorded No additional effect beyond the effects described in 'No retraining' Personalized recommendations The more events streamed, the more personalized the recs become. Impression discounting will occur for cold start items when new users records include impressions data.
putEvent with new user Personalized Ranking Personalization begins after 1 event, but will be more visible after ~2-5 events with 1-2 seconds delay after the PutEvents call after each event is recorded - Personalized recommendations Using Personalized Ranking, it is more difficult in many cases to see the direct impact of putEvents records as a customer supplied curated list is re-ranked (compared to user-personalization where recomendations are generated from the full vocabulary of items in the catalog based on the learned model behavior/metadata features and user interaction history.)
putEvent with new user SIMS - - Included in the model to generate the recommendations SIMS doesn't really do personalization so in the context of new users being added with PutEvents, a new user's events are considered in similar item recs only after retraining.
putUser User personalization - - Personalized recommendations Users added via putUser will be warm users based on the conbination of their known interaction history and userID after the next full retraining.
putUser Personalized Ranking - - Personalized recommendations Users added via putUser will be warm users based on the conbination of their known interaction history and userID after the next full retraining.
putUser SIMS - - No effect SIMS doesn't really do personalization so in the context of new users being added with PutUsers, a new user's events are considered in similar item recs only after retraining.
putItem User personalization - Appear as cold start items eligible based on the exploration age cut-off when exploration is enabled. Personalized recommendations For new/cold start items, recommendations are personalized based on the user's interaction history and item metadata of new/cold start items. Cold start items (eligible based on the exploration age cut-off when exploration is enabled) will be included during the next update. Cold start items will be auto-updated based on impression discounting from interacting generated during exploration. This weighting is non-linear combined with metadata based features but cold start items which are less popular (provided in the impressions field via putEvents will receive less exploration weighting over time)
putItem Personalized Ranking - - Personalized only after some interactions -
putItem SIMS - - New interactions are included in the model to generate similar item recommendations based on co-occurence SIMS doesn't really do personalization so in the context of new users being added with PutEvents, a new user's events are considered in similar item recs only after retraining.

Technical Enablement Links:

  1. Overall Samples: https://github.com/aws-samples/amazon-personalize-samples
  2. Getting Started: https://github.com/aws-samples/amazon-personalize-samples/tree/master/getting_started
  3. POC in a Box 2.0: https://github.com/aws-samples/amazon-personalize-samples/tree/master/next_steps/workshops/POC_in_a_box
  4. Use Case Based Notebooks: https://github.com/aws-samples/amazon-personalize-samples/tree/master/next_steps/core_use_cases
  5. Data Science Tools: https://github.com/aws-samples/amazon-personalize-samples/tree/master/next_steps/data_science
  6. MLOps for Personalize: https://github.com/aws-samples/amazon-personalize-samples/tree/master/next_steps/operations/ml_ops
  7. Monitoring/Alerting/Cost Optimization: https://github.com/aws-samples/amazon-personalize-monitor
  8. Personalized Marketing Outreach: https://aws.amazon.com/blogs/architecture/amazon-personalize-customer-outreach-on-your-ecommerce-platform/

Demos/Workshops:

Technology Partners:

There are several technology partners that provide complementary functionality to Personalize that can accelerate customers getting to production with Personalize or enhancing the ROI on implementing personalization with Personalize.

Customer Data Platforms - Event Collection/Activating Recommendations

Segment is a Customer Data Platform. They are an AWS Advanced Technology Partner and hold the Digital Customer Experience (DCX) and Retail competencies.

Segment helps customers with Personalize in the following ways:

  • Event collection - this is a core capability of Segment. Customers use Segment to collect clickstream events across their web app, mobile apps, and other integrations. These events are collected, validated, and fanned out to downstream destinations configured by the customer. One of these destinations is Amazon Personalize.
  • Customer/user profile identity resolution - because Segment sees events across all channels for a customer's users, they are able to create a unified customer profile. This profile/identity is key to being able to deliver omnichannel personalization.
  • Activation across an organization's other marketing tools - because Segment allows customers to create connections to other marketing tools, attaching personalized recommendations from Personalize to profiles in Segment allows customers and downstream partners to leverage those recommendations in their tools.

Resources

mParticle is a Customer Data Platform. They are an AWS Advanced Technology Partner and hold the Digital Customer Experience (DCX) and Retail competencies.

mParticle helps customers with Personalize in the following ways:

  • Event collection - this is a core capability of mParticle. Customers use mParticle to collect clickstream events across their web app, mobile apps, and other integrations. These events are collected, validated, and fanned out to downstream destinations configured by the customer.
  • Customer/user profile identity resolution - because mParticle sees events across all channels for a customer's users, they are able to create a unified customer profile. This profile/identity is key to being able to deliver omnichannel personalization.
  • Activation across an organization's other marketing tools - because mParticle allows customers to create connections to other marketing tools, attaching personalized recommendations from Personalize to profiles in mParticle allows customers and downstream partners to leverage those recommendations in their tools.

Resources

Analytics/Measurement/Experimentation

Amplitude is an AWS Advanced Technology Partner and holds the Digital Customer Experience (DCX) competency. Amplitude helps customers with Personalize in the following ways:

  • Product insights - Amplitude provides visibility into the event types that lead to conversion through sophisticated funnel analysis. This provides the insight that customers need to optimize their event taxonomy and select the right events and metadata fields to train models on in Personalize.
  • A/B test evaluation - Amplitude provides online measurement of A/B tests that can be represented by personalized customer experiences powered by Personalize.

Resources

Optimizely is a market leading A/B testing platform. They are an AWS Advanced Technology Partner and hold the Digital Customer Experience (DCX) competency.

Optimizely helps customers with Personalize in the following ways:

  • A/B test results - a core offering of Optimizely is measurement and reporting of experiments such as personalization techniques.
  • Feature flagging - enable/disable personalized experiences

Resources

Messaging

Braze is a market leading messaging platform (email, push, SMS). They are an AWS Advanced Technology Partner and hold the Digital Customer Experience (DCX) and Retail competencies.

Braze helps customers with Personalize in the following ways:

  • Deliver personalized messages to customers on the right communication channels through either a real-time integration or batch integration.

Resources

Direct/Managed Integrations

Magento 2: A Magento 2 extension was developed by a Magento and AWS partner called Customer Paradigm. The extension was NOT developed by Adobe Magento.

The extension can be easily installed into any Magento 2 storefront whether it's running on-premises, another cloud provider, or on AWS. Amazon Personalize is always accessed in the customer's AWS account.

Resources

Shopify: Obviyo (formerly known as HiConversion) has built a managed integration with Personalize for Shopify storefronts. This means Obviyo is managing Personalize in their AWS environment and Shopify merchants pay Obviyo for personalization capabilities that are powered by Personalize.

Resources