How Spotify & Snap Use ML for Recommendation Systems

The speakers discuss the challenges and future of recommendation engines, including the importance of data, privacy, and explainability.

• Online adaptive learning is important for quickly adapting to user needs
• Great data is necessary for building a machine learning recommendation engine
• Understanding user journeys and causal impact is important for effective recommendations
• Long-term metrics, predictive metrics, and short-term metrics are all important for measuring success
• Privacy and fairness are ongoing challenges for the ML community
• Explainability is important for user control and understanding of recommendations

The speakers mention the importance of understanding user journeys and causal impact in order to make effective recommendations. They give an example of predicting the LTV of a user over a 60 month horizon through a predictive model, which takes into account various inputs and local features. They also discuss the importance of explainability, citing Netflix's use of associated learning to explain recommendations.

Content comes in many forms—from traditional media with professionally developed content to social media with user-generated content, and everything in between. Over the past decade, media platforms have evolved significantly to keep up with the pace of newly developing content and content trends. For example, social media now support multiple forms of content, including images and short-form video, and traditional media platforms have expanded to support in-house content and streamed content. This expert panel will explore the importance of recommendation systems, discuss the importance of the data that powers these systems in the ever-evolving media platform experience, and cover how and why recommendation systems are so critical for growing media-based businesses.