Senzari, a cutting-edge entertainment data and technology company, announced at the 2015 DEW Expo the upcoming launch of MovieGraph, a new entertainment graph services platform that uses machine learning and graph technology to develop detailed semantic graphs around video content. MovieGraph connects and extracts knowledge from large sets of TV show and movie industry data which allows the platform to provide personalized search and recommendation services. MovieGraph data and functionality can be integrated with third-party applications using the upcoming MovieGraph API.
We reached out to Demian M. Bellumio, Senzari Chief Operating Officer, who spoke to ProgrammableWeb about the technology behind MovieGraph and some of the development aspects of the new platform. Bellumio explained to ProgrammableWeb that Senzari used machine learning to automate the process of mining data from publicly available movie and TV show information sources. Bellumio said that MovieGraph required less content to build than MusicGraph, Senzari's music data and intelligence platform. MusicGraph mines more than 7 billion data properties; The mined data along with graph technology, powers MusicGraph's song recommendation, graph search services and complex graph analytics.
Bellumio told ProgrammableWeb that songs contain far more metadata and metrics than movies and television episodes which is why Senzari focused on metadata while building MovieGraph. He also said that Senzari trained machine learning algorithms to break down the narratives of movies, extracting the data with precision across each element. The company designed their own matrix for cataloging movies; MovieGraph uses machine learning techniques to semantically tag and organize every movie and TV show across hundreds of dimensions. Senzari also added proprietary narrative features to MovieGraph such as setting, conflict, symbols or tones present in a film. Using automated machine learning algorithms greatly reduced the development time of MovieGraph, which took only a few months to build; It took Senzari two years to mine and semantically connect the data contained in MusicGraph.
MovieGraph uses graph technology to break down user preferences which allows for more precise connection points and dimensions. This precision makes it possible for MovieGraph to provide greater personalization and a powerful Netflix-like recommendation experience for application end users. Bellumio told ProgrammableWeb that Senzari is "really happy with the results and user experience of MovieGraph."
The MovieGraph platform features an API that developers can use to add a very personalized Netflix-like recommendation experience to their applications. While the primary use case for MovieGraph is personalized recommendation engines, MovieGraph's vast repository of movie and television data along with graph technology allows for many other types of use cases. MovieGraph is capable of performing advanced graph search queries such as "In the past five years, what movies have won Oscars with a female lead character and also have been liked by my friends on Facebook?"
The MovieGraph platform works with MusicGraph to form connections between video and musical works. Content creators can use MovieGraph to find suitable music content for their works along with information about the music creator. Movie and television series creators can use MovieGraph to find similar music from independent artists to use as alternatives to music from major record labels or major artists. The use of music from independent artists is generally less expensive than music produced by major record labels.
Bellumio explained to ProgrammableWeb that one of Senzari's goals when it comes to MovieGraph is to provide companies and developers a way to set up and implement a Netflix-like recommendation engine without a huge expense or long development time. He also said that most of the television and movie databases available today provide only metadata and a basic API (or don't provide an API). The MovieGraph API not only provides access to movie and television metadata but also provides access to the platform's graph search functionality. MovieGraph is designed to be a powerful entertainment data and intelligence platform, not just another movie metadata database.
The upcoming release of the MovieGraph platform and API is well timed considering that Netflix shutdown its public API this past November. MovieGraph should prove to be a viable option for developers who had built applications with the now defunct Netflix API. Bellumio is quoted in the press release as saying:
"The music industry is hacker-friendly with an enormous developer community. We've become a trusted partner by embracing that 'share-break-build' mentality. Most movie intelligence solutions in the market have a restrictive, 'black box' approach, sharing limited data while forcing a complex and costly sales integration process. The MovieGraph demo (www.moviegraph.com) demonstrates that a Netflix-like experience can be created by leveraging our API, allowing for a more personalized experience than that of the market leader, and requiring absolutely no interaction with engineering or sales support."
The MovieGraph platform and API is scheduled to be released in March 2015. Visit Senzari.com for more information about the MovieGraph and MusicGraph platforms.