Enterprise 2.0 Meets the Semantic Web

By SemanticWeb.com on June 10, 2009 3:26 PM

Jennifer Zaino
SemanticWeb.com Contributor

Discussion is growing about the relationship between the semantic web and the 2.0 Enterprise. Expect to hear more about it at next week’s 2009 Semantic Technology Conference in San Jose from speakers such as Alexandre Passant, a postdoctoral researcher at the Digital Enterprise Research Institute (DERI) at the National University of Ireland in Galway.

In many enterprises today you’ll find tools such as blogs and wikis — it’s part of the infusion of social media into the workplace. A significant problem, however, is the difficulty employees have in finding the content they need from among these corporate sources — currently a labor-intensive manual task — and interact with them in more effective ways.

“Semantics provides better interoperability between these applications, and on top of that, provides new services — mashups that combine data from different sources,” says Passant. According to Passant, enterprise 2.0 can be used to foster collaboration and social intelligence, but issues raised include:

Information fragmentation: Among various wikis, blogs, or other content management forums, the same object may be discussed in various different ways, leading to confusion and disconnections.

“When you add a semantic layer you then describe the content in Wikis or blogs using the same data format, and that then lets you plug in some search to actually find if there is a certain topic in a wiki or blog post, and then you can provide a user interface on top of that,” he notes.

The document/data gap: Wikis play host to lots of valuable information but in plain text content. You can’t query a Wiki engine to, for instance, find all Wikis related to a certain type of project, meaning you have to dig around to find what you need. “Instead of representing that information in plain text, you need some structured information. Semantic wikis are extensions to the Wiki platform, so using a special Wiki syntax you can create a structure of information,” Passant notes. That leads to the ability to make precise queries.

Tagging: Simple keyword tagging leads to a heterogeneity of tags. Putting a semantic layer on top with a unique identifier enables reference to the same thing even when different text is used. That is a help in finding related content, too. “If you move to a semantic layer and get structured information, then you’ll be able to find connections,” he notes.

Concerns about the complexities of semantic technology to support data interoperability and mashups that rely on that is lessening, he thinks, as the semantic infrastructure is growing. “It was quite difficult a couple of years ago to store scalable semantic data, but now there are some solutions to provide this,” he says. “So that reduces costs and shows people it works and can be implemented.”

Other drivers of support for semantic web-Enterprise 2.0 integration include growing interest in the Linking Open Data project; the interest of leading search engines such as Google and Yahoo in indexing semantic data on the web; and the fact that providing a semantic web layer doesn’t mean throwing away existing systems, but rather adding plug-ins on top of existing tools.

“Users don’t have to learn new tools and change habits,” he says — except, perhaps, for the habit of not necessarily being willing to share information in the first place. But that’s more of an Enterprise 2.0 issue than a semantic one, he notes, which can and should be addressed through privacy settings and policies that deliver information on who published data, user rights for accessing data, and so on.

Comments