Gracenote assigns mood elements in descriptive metadata.
Gracenote, subsidiary of video audience measurement and analytics group Nielsen, has released the first in a series of metadata-based products for content search and recommendation under its Advanced Discovery brand.
The firm is trumpeting the use of machine learning algorithms to derive contextually relevant metadata from content going beyond traditional basic descriptors such as genre and title, to include mood, theme, scenario and characters. The firm has kicked off with the first product in the new series called Gracenote Video Descriptors, designed to obtain a finer grained and deeper description of content as a foundation for more personalized video selection.
Gracenote cited as an example Game of Thrones which could be categorized under themes such as “Greed” and “Betrayal” and then Scenarios including “Power Struggle” and “Manipulation.” It could also be assigned to Mood elements like “Dark” and “Gripping”, while the characters could be classified from “Royalty” to “Dragons.”
So, under Scenarios, Gracenote data could be used to identify similar programs based on related descriptors such as “Sweet Revenge” or “Good versus Evil.” The firm contends that by understanding better what Game of Thrones is about and relating those attributes to other programs, Gracenote Video Descriptors can enable more deeply nuanced discovery that unearths content most likely to resonate with individual viewers.
It remains to be seen how successful that approach will be compared with conventional more superficial search methods, but it is notable that Gracenote does not believe machine learning can yet be relied on to extract the metadata unaided. The firm will still rely on a combination of its data scientists and editorial teams to oversee the process with machine learning operating almost in a subsidiary role as a tool to automate some aspects of the metadata generation.
Further releases from the Advanced Discovery suite will be rolled out in 2019, while Gracenote was demonstrating Video Descriptors alongside other entertainment data products at CES (Consumer Electronics Show) 2019 in Las Vegas.
You might also like...
New England Patriot quarterback, Tom Brady, entered Mercedes Benz stadium in Atlanta, GA on February 3rd having already won five Super Bowl games. And through four-quarters of play, all delivered by a television crew of hundreds of technicians, sports casters…
This FREE to download eBook is likely to become the reference document you keep close at hand, because, if, like many, you are tasked with Preparing for Broadcast IP Infrastructures. Supported by Riedel, this near 100 pages of in-depth guides, illustrations,…
In their latest hyper-realistic VR weather warning, The Weather Channel helps viewers better understand the potential dangers created by ice storms.
As broadcasters continue to successfully migrate video and audio to IP, attention soon turns to control, interoperability, and interconnectivity to improve reliability and efficiency. In this article, we investigate IP control as we move to IP infrastructures.
Thanks to improved streaming technology, a lot more fans are going to be watching the Super Bowl on mobile screens.