Jennifer Zaino
SemanticWeb.com Contributor

The recording industry, once an Internet laggard, isn’t about to be put in that position again. And semantic web technology is helping on that score.

The technology is helping firms — including Interscope Records (Universal), Independent Label Group (WMG), RCA Music Group (Sony) and entertainment services company The Orchard — take a proactive role in understanding music activity across the web and use that information to better market artists and their work.

“I think that at this stage in the game the idea that major labels are laggards and don’t get this [web] stuff is not accurate at all,” says Jim Lucchese, CEO of The Echo Nest, which is providing the Fanalytics platform to these companies. “Every major label has really smart, empowered digital groups now who understand the music world as well as anyone on the web.”

One thing those groups understand — which has made it possible to sell them on The Echo Nest’s service — is that the old ways of promoting to radio networks and their established formats don’t work in the world of the Internet. In contrast to the more stable radio industry, where formats don’t change overnight, on the web new music blogs and writers pop up every day.

With two years — and counting — of web crawling in its background, Fanalytics applies semantic technology to extract meaningful descriptive terms from unstructured data about artists and tracks being written about: That, for example, a new blog post applies ‘”funky” as an important term for James Brown; such a descriptive is then used to culturally understand something about how the music world is described, and is pulled out of thousands of reviews a day. Similarly, Fanalytics essentially “reads” every music writer on the web today and then builds a case profile that extracts entities descriptive of the posters’ interests.

“We don’t just understand keywords but their musical tastes,” says Lucchese.

In the same way that Fanalytics can match up artists and cluster music tastes based on the data about the descriptives applied to them and their genre, it can pull data from its thousands of writer profiles that help recording companies discover those most likely to respond to campaigns about particular artists. That way, Lucchese says, companies can really target their message and engage with the blogger, rather than just sending out hundreds of hit or miss blasts.

“It would be impossible to keep up without semantic web technologies,” he says. “We as a company deal with these massive data problems that digital music creates, problems that you can’t keep up with manually. New blogs and sites are coming up and disappearing every day. Trying to keep up with that from a manual research team is virtually impossible. You have to do web crawling to keep up and use semantics and machine learning technology to process and synthesize that information and stay current.”

Lucchese expects to announce in the next couple of weeks a couple of other major labels signing on. It also has in the works adding sentiment analysis of online writers’ work to the picture.

“In the first couple of years we pulled out descriptives to understand, for example, that James Brown is funky. We now apply different math to understand in real time, of the hundreds of blog posts about a particular artist, how many are positive or negative,” he says.

That is currently in alpha testing as part of a small business innovation research grant the company won from the National Science Foundation. Lucchese says his team wasn’t sure if record labels would be interested in sentiment, but it turns out they are. “They look at the web as a powerful early warning system or indicator for marketing efforts,” he says. They want to be able to use online pre-release indicators of videos pushed to YouTube, for example, to gauge potential success of video or single launches. “If you can capture [the sentiment] data [around these], they can use that data to inform how to market an artist, regionally or where on the web, or how much money to put behind him, and how to sell that artist to radio or video.”

Future plans from The Echo Nest are deeper tools to help artists and labels better understand who their fan base on the web is. Meantime, the company is working through its pricing model. It’s selling direct to the major labels and plans to use distribution models to offer lower tier access to data at a realistic price point for potential clients such as individual artists who want to promote their work.

Semantic Tech & Business Conference Returns to San Francisco

Semantic Tech & Business Conference returns to San Francisco in June! Join us from June 3-7 for complete coverage of Big Data, Linked Data, Extreme Information Management, and Semantic Web. From breakthrough approaches to solving business problems to the big data implications of fast–evolving technologies, SemTechBiz provides you with an unparalleled interactive experience and delivers tangible business value. We're offering a special early rate when you register by February 17. Sign up now!