Video Breakthroughs

In part one of this blog post, we detailed the different components of Netflix personalization. We also explained how Netflix personalization, and the service as a whole, have changed from the time we announced the Netflix Prize.The $1M Prize delivered a great return on investment for us, not only in algorithmic innovation, but also in brand awareness and attracting stars (no pun intended) to join our team. Predicting movie ratings accurately is just one aspect of our world-class recommender system. In this second part of the blog post, we will give more insight into our broader personalization technology. We will discuss some of our current models, data, and the approaches we follow to lead innovation and research in this space.

The goal of recommender systems is to present a number of attractive items for a person to choose from. This is usually accomplished by selecting some items and sorting them in the order of expected enjoyment (or utility). Since the most common way of presenting recommended items is in some form of list, such as the various rows on Netflix, we need an appropriate ranking model that can use a wide variety of information to come up with an optimal ranking of the items for each of our members.

Late last week, video search and recommendation technology provider Digitalsmiths quietly introduced its "Seamless Discovery" platform, targeted to pay-TV operators, consumer electronics companies and content producers who want to deliver highly relevant recommendations to their users. The platform addresses the urgent problem that users are fragmenting their viewing over multiple devices where the discovery experience is inconsistent and lacking.

In a phone briefing, Digitalsmiths' CEO Ben Weinberger explained that a key differentiator for Seamless Discovery is that it draws from multiple data sets in order to provide recommendations, resulting in improved relevance. At the core is metadata Digitalsmiths creates on the programming available from the pay-TV operator, CE device or content owner. For pay-TV operators specifically, this involves ingesting full schedule information from sources like Tribune Media Services. This metadata is mapped with contextual and behavioral data and "social graph" information from Facebook along with other inputs. The system learns over time from the choices the user makes which of these factors is most relevant, tweaking future recommendations accordingly.

Jinni has announced an enhanced version of its API, letting TV and device makers add more personalization and social recommendation features to their movie and TV show guides. Jinni, a Tel Aviv, Israel-based company, creates search and discovery recommendations based on the viewer's mood and taste.

Capitalizing on one of the biggest trends in online video, the API now offers social recommendations from friends that also match the viewer's profile. Jinni doesn't rate all friend recommendations equally, but looks for taste matches. The company previewed this feature at CES, and is now releasing it.

Jinni can now recommend programming from live TV, a big step for a product that previously only offered video-on-demand and over-the-top content recommendations.