Music data startup Senzari unveils new engine for better music analysis

Frank Sinatra - Generic

Senzari, the big data startup that specializes in music, on Monday released its web-based graph analytics engine dubbed MusicGraph.ai. The goal is to market the product to streaming music companies like Pandora and record labels. The new engine is essentially the dashboard and backend system that users can access to get detailed reporting on their accounts with MusicGraph, a database that the company claims contains over a billion pieces of music data — including song lyrics and the songwriters behind the year’s top hits.

While other music recommendation services like Gracenote maintain their own vast libraries of music-related information, what makes Senzari stand out is its new engine’s ability to run custom analytics and machine-learning algorithms for users to discover new ways of interpreting the data, said Senzari chief operating officer Demian Bellumio.

For example, using the MusicGraph.ai, customers can perform complex types of analysis like triangle counting, in which the system can look at all artists that a user wants analyzed and from there determine how artist A is connected to artist B who is connected to artist C, and so on. From this type of analysis, Bellumio said, a user can learn who is the most connected musical artist of all time (it’s apparently Frank Sinatra, by the way, according to a demonstration of the engine by Senzari).

Graph displaying how connected Frank Sinatra is

Graph displaying how connected Frank Sinatra is

Senzari collects music-related data from a multitude of databases and includes social data as well, such as how many plays Adele has on Last.fm and her number of Facebook followers. Using this information, Sensari’s customers will be able to run either analytics or machine-learning algorithms on top of the particular pieces of data they want analyzed in an interface that resembles a simplified content management system.

The analytics algorithms will let customers discover information like what type of songs are particular demographics of users listening to, while the machine-learning algorithms can unearth information like how do certain types of lyrics influence how well a song can perform on the music charts.

Display of Music.Graph.ai engine

Display of Music.Graph.ai engine

With the MusicGraph.ai engine, Senzari is pretty much handling all of the leg work involved with performing the data analytics work on top of the music data, stored in Amazon Web Services. Potential customers still need to purchase AWS storage, but they won’t be responsible for performing the computations, thus eliminating the need to hire costly database administrators, said Bellumio.

“Running 100 machines and deploying an algorithm — you need a team to do it and it is costly,” said said Bellumio. “This takes away from the equation.”

Post and thumbnail images courtesy of Shutterstock user Northfoto.

loading

Comments have been disabled for this post