Directed Edge Launches Recommender System for the Social Web

Share Article

Directed Edge announced today the initial public beta of its recommender system, an engine for finding related media, people and events on the web. The Wikipedia-based demo shows how graph analysis of connection structures on the social web can be used to deliver relevant recommendations to users.

Directed Edge announced today the initial public beta of its recommender system, an engine for finding related media, people and events on the web.

The Wikipedia-based demo shows how graph analysis of connection structures on the social web can be used to deliver relevant recommendations to users.

"We believe that Directed Edge's technology is the key to finding information that really interests users," said Willi Ruopp, Managing Director of the online music service musiqtogo. "It's about discovering, not searching."

Automatic recommendations have become an important part of popular web destinations, driving sales in online marketplaces by suggesting similar books, movies or music to users and driving higher traffic in news sites by displaying related stories.

Directed Edge's engine makes it possible for partner sites to deliver recommendations to their users without developing that technology in-house by offering a web service to its customers that allow recommendations to be integrated directly into their sites.

In the technology preview released today, Directed Edge has built upon the freely-available data from Wikipedia to show users related articles for every page in the encyclopedia. This makes it easy to quickly explore any given topic and discover new related topics. Scott Wheeler, technology lead of Directed Edge commented, "It's a nice way to show people what our engine is capable of and what it can do for customers' sites. It's a big first step for us."

Directed Edge uses a "graph" - or connection-based - approach for finding similar pages on partner sites or the web. Unlike most recommendation systems, which are based primarily on users' product ratings (collaborative filtering) or text analysis (indexing or semantic analysis), Directed Edge's algorithm can take advantage of the emerging social elements of modern web destinations to find related content. For instance, it can recommend music based on a user's friends or events based on a user's business contacts and professional memberships.

In addition to the encyclopedia-based demo, during its beta phase Directed Edge will begin integrating with several pilot customer-sites from different internet sectors.

"It's a very exciting technological area - we knew at first sight that we could make great use of it," said Sven Herschel from Loge2, the Berlin-based social network for culture and entertainment. "I believe their graph analysis can really make an impact on our site and we're looking forward to working with them."

###

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Valentin Hussong
Directed Edge
+49 30 290079228
Email >
Visit website