AWS Media Blog
Create personalized user experiences with ThinkAnalytics and Amazon Personalize
This blog is co-authored by Nilay Naik from ThinkAnalytics and Kevin Yao from AWS.
The ThinkAnalytics Think360 platform helps media, entertainment, news, and direct-to-consumer video providers use a data-driven approach to increase revenue and customer engagement. The platform integrates Amazon Personalize from Amazon Web Services (AWS) to allow a wide selection of personalized recommendation solutions running on Think360 in the AWS Cloud. The Amazon Personalize service lets developers quickly build and deploy curated recommendations and intelligent user segmentation at scale using machine learning.
This integrated solution from ThinkAnalytics provides data-driven management tools that enable a complete personalized TV solution – without the need for additional machine learning (ML) or development resources. Business managers have the power to activate an advanced content discovery platform optimized for the media and entertainment industry. Integration with Amazon Personalize extends customer choice, giving service providers even more options of the recommendations techniques that best suit their requirements for each user experience, screen, or device while retaining business control over the solution.
The Think360 content discovery and viewer engagement platform is an AI/ML-based personalized search, recommendation, and user engagement solution that puts control of content discovery in the hands of the business, revenue, and product owners. Personalization of the viewer experience is proven to deliver increased viewing, purchasing, higher satisfaction, and perceived value for money, leading to reduced churn.
ThinkAnalytics integrated its platform with Amazon Personalize and automated the process of training the Personalize ML models. ThinkAnalytics ensures the highest standards of customer deployment with a team of AWS Certified consultants who have Amazon Personalize expertise. The combination of AWS services and Think360 Content Discovery Platform makes extensive use of: Amazon Personalize, Amazon EC2 , Amazon RDS, AWS Lambda, Amazon Simple Storage Service (S3), S3 Glacier, Amazon VPC (NAT Gateway, Elastic Load Balancing), Amazon Elastic Block Store (EBS), Amazon CloudWatch, AWS CloudTrail, Amazon Route 53, and AWS Fargate.
The integration allows customers to take full advantage of the latest ML techniques, massive operational scalability, and AWS services. Deploying on AWS ensures support for peak loads at scale that are required in the media and entertainment sector, while minimizing costs.
Boost and amplify AWS Personalize with Think360
Think360 accelerates development of AWS Personalize campaigns by providing an abstraction layer for instrumenting recommendations and automating model creation and training. Leveraging the data architecture of the Think360 platform, training data is captured, and model training tasks are automated using AWS Fargate scheduled tasks. This simplifies the data cleansing and integration tasks that would typically be required, allowing service providers to concentrate on the quality of recommendations.
High-level overview of ThinkAnalytics Think360 and Amazon Personalize integration
Viewer activity such as playing, downloading, rating, bookmarking, etc. (over 20 different pre-defined learning activities are supported along with the ability to configure new ones) are delivered to Amazon S3 storage automatically by Think360. AWS Fargate is then used to trigger timely auto-rebuild/refresh of the Amazon Personalize campaigns.
The ThinkUX platform supports serving recommendations using pre-built models within Amazon Personalize, providing the option to add Amazon Personalize as a recommendation use case. When selecting it, the user has the option to select a campaign from a list of data-driven campaigns, via simplified API calls.
The ThinkUX platform lets service providers integrate use cases that use AWS Personalize campaigns along with a multitude of other recommendation types via a drop-down list. The abstraction layer that ThinkUX provides allows operators to switch or combine campaigns and A/B test campaigns without additional development overhead. ThinkUX allows for rapid development, deployment, and evaluation of new personalization use cases. It also allows for effective management, control, and prioritization of UI use cases, content recommendation types, and parameters. Use cases are easily defined, edited, and published to provide powerful new user experiences. Development resources are not required to implement an improved user experience that drives more viewing and engagement. Business owners can implement changes in minutes, including complete A/B testing to verify the effectiveness of the new experience before publishing it to the entire viewer base.
The ThinkAnalytics Think360 platform is available in the AWS Marketplace and can be deployed in a provider’s AWS instance or as a fully managed ThinkAnalytics SaaS service. To learn more, contact ThinkAnalytics at firstname.lastname@example.org or connect with your AWS account representative.
ThinkAnalytics is an AWS Partner for media and entertainment specializing in content discovery, viewer analytics, and contextual advertising solutions that enable video service providers, studios, broadcasters, and media companies to deliver personalized experiences to their customers. The company has delivered content discovery and viewer insights to over 80 service providers serving approximately 450 million subscribers in 43 languages with 7 billion recommendations per day.