Video: Demo: How to Make Your Video More Searchable

Learn more about OTT personalization at Streaming Media's next event.

Watch the complete video of this panel from Streaming Media West, DT202B: The Path to Personalization, in the Streaming Media Conference Video Portal.

Read the complete transcript of this clip:

Mark Tubinis: People search for things in different ways. They'll use different terms. It’s all based on how they think. If your metadata is not enriched and that person does a search, either a voice search or enters a search term, and that information is not part of the metadata, that consumer is not going to find what they want.

The only way to do that is just to continue to add metadata to the information that improves search results. The information that comes in from the variety of sources of content is often not the same, it's not aligned. Simple things like the rating might be actually formatted differently when you get it sourced from one source versus another.

When you get into the content management challenge, you've got to make sure that you're properly rooting out content that's inappropriate for youngsters and if they have not marked the content in the same way then you have a problem. So we have to normalize this metadata, as well.

But the most important thing is the ability to enrich. So add metadata to the information so that you have a better chance of doing better recommendations for those end users and also making the search process easier for the end customer.

I'd like to show you an example of something we're doing with our content management system that allows us to add that enriched metadata by actually analyzing using AI tools. We're analyzing the actual video source and extracting that metadata directly from that analysis.

This is a view of our content management system and we're going to show a title called Tears Of Steel, whereby you can see some of the metadata that's part of it.

Down at the bottom there's some enrichment information that we've actually articulated that we want the AI engine to go find. Working with a company called Dive.tv, which is a European AI company. They’re actually going through and analyzing the video and identifying different very, very specific locations, as an example. They can go in and look for fashions and even brands of fashions. They can go in and look for brands of automobiles. They can go in and basically add all this information frame-by-frame into the system.

We can then import that information into our content management system so it can be used for both recommendations and for discovery. And there are a lot of other use cases of this technology that we're finding that people are willing to pay significant amounts for.

This is an added feature to our content management system and we're looking at other use cases that we can have. So here it's just demonstrating that we've gone into our content management system so this is now part of that big warehouse of information that we have. We see that we've marked the Amsterdam Central Station. And who knows, somebody may have gone on their honeymoon to Amsterdam and had some great dinner near the Central Station and they might be on their anniversary and they want to go look for that, and it turns out that there's a movie called Tears Of Steel that they can go watch. It may not be appropriate for a romantic interlude.

This is just an example, a zoom-in on that information. There are other capabilities that this AI engine gives us. For instance, establishing the mood of a particular part of the video.

As an example, there's a really exciting car chase that includes Audis. There's also a set of scene changes that it has marked here so that after you've put the content in and you have the rights to be able to do ad insertion.

Think about the use case where you may have information about the end user that they happen to be an Audi lover, and you just had this exciting Audi car chase in the Transporter movie, of course. The end result is you can take the scene break right after that car chase and present an Audi commercial to them from the ad system.

Streaming Covers
Free
for qualified subscribers
Subscribe Now Current Issue Past Issues
Related Articles

Video: The Challenge of OTT Personalization

SeaChange International SVP, Strategy Mark Tubinis discusses the difficulty of managing the data necessary to create personalized OTT experiences in this clip from Streaming Media West 2018.

Video: How Personalization is Revolutionizing OTT

Nice People at Work's Laura Delcor and Google Cloud's Luke Dyer discuss the impact of personalization and interactivity on the OTT industry in this clip from their Live Streaming Summit panel at Streaming Media West 2018.

Video: How Reinforcement Learning Enables Personalized Viewing Experiences

Microsoft Principal Product Manager Rafah Hosn makes the case for reinforcement learning as a machine learning paradigm for content personalization in this clip from Streaming Media East 2018.

Video: Pros and Cons of Supervised Machine Learning for Content Personalization

Microsoft Principal Product Manager Rafah Hosn discusses the benefits and limitations of a content personalization strategy based on supervised machine learning in this clip from Streaming Media East 2018.

Video: How to Use Machine Learning to Create Personalized TV Experiences

ZoneTV's Tom Sauer describes how machine learning can be used to overhaul the TV world and deliver more individualized experiences in this clip from Streaming Media East 2018.

Video: What Does Personalization Mean in OTT Video?

Sea Change International VP Marketing Kurt W. Michel discusses analytics- and AI-driven personalization in this clip from his presentation at Streaming Media East.