“The Semantic Web promises to organize the world’s information in a much more logical way,” says Marc Fawsi, an analyst at Evolving Trends in San Francisco. “And once machines can understand and use information, the world will never be the same.”
Semantics in Action
Today, Semantic Web technology is seeping into corporations and government agencies to build sophisticated knowledge bases and knowledge management systems, says Nova Spivack, CEO of Radar Networks, in San Francisco. “Government agencies as well as firms like Eli Lilly and Oracle are using it to improve discovery of what exists in unstructured documents and data collections scattered through their organizations.”
For example, say a pharmaceutical company wanted to find all information related to a certain disease, including all the drugs and treatments and conditions related to that disease. Previously, a search would bring up only those documents or database records that contained the keywords specified by the human searcher. With the Semantic Web, however, the machine could make inferences: because a certain drug is commonly used to treat that disease, or because a specific symptom is typically associated with that disease, any information that contained references to those drugs or symptoms would also be retrieved, even if the disease wasn’t mentioned by name.
A Bit of a Disconnect
Many corporations are already making use of Semantic Web technology — but often management isn’t aware of it because it is being used by employees in unauthorized ways, according to Scott Abel, CEO of Indianapolis-based The Content Wrangler. He points to the ways that some enterprise customer call centers are beginning to use tagging services like del.icio.us to help them better organize their reference materials. “Call center representatives are often challenged by having to look through massive amounts of online resources provided by the corporation to answer customer questions. This can include technical manuals, training manuals and lists of frequently asked questions (FAQ),” says Abel.
Unfortunately, the way these resources are organized doesn’t really correspond to the way the employees must use them. But by “tagging” various pieces of this online content with such things as the name of the customer who needed it, or the specific problems customers were having that were solved by it, the call center can begin to compile a knowledge base that corresponds more closely to organizing the information the way its employees use it. “For example, a customer representative can do a quick retrieval of all the specific instances when a particular customer called in to complain about the battery life of a product,” says Abel.
This helps the corporation in many ways, “not the least of which is that when a customer calls in, the representative is likely to be able to respond more quickly and accurately,” says Abel.
Mixing and Mashing Disconnected Sources
Another example of corporate use of the Semantic Web involves “mash ups,” or the act of bringing together and consolidating information from different online sources into a single integrated experience. For example, a technology company could mash up internally generated documentation with information provided by customers on a user forum to come up with a more comprehensive — and user friendly — source of data about products for customers. “In these forums, you have users helping users, and it adds up to these huge libraries of valuable information that companies can leverage to better serve all customers,” says Abel. “The Semantic Web allows companies to integrate this information much more easily than a traditional database could.”
Agrees Spivack, “You could think of the Semantic Web as a new kind of middleware in which you can do data integration without having to go through application integration.”
For now, most corporations are making only limited forays into the Semantic Web. For example, many firms are using Semantic Web technologies internally but confine their implementations to information stored within the corporation, rather than venturing out onto the Web itself. “This turns out to be much easier than what people are struggling with in the larger world, because they have control over all the elements and don’t have to worry about all the different technologies and standards that currently are competing with each other on the Web,” says Fawsi. However, he predicts, “As standards begin to become established, that will change.”