There’s a fun comparison app on FindtheBest.com, the new venture from Kevin O’Connor, the mind behind DoubleClick. FindtheBest is a comparison engine that lets users find a topic (business schools, eBook readers, teeth whitening strips), compare options up against each other (just how much hydrogen peroxide is in a particular whitening system anyway?), sort and narrow data with smart filters, and make a decision based on what they discover.

That includes helping them determine, with a side-by-side chart comparison, which Q&A site might best serve their needs – which is how you get to compare FindtheBest with some 40 other options, including ones we’ve discussed here, such as TrueKnowledge and Swingly. For the Q&A comparison app, you can see things like which has the best Alexa ranking; how each site generates answers; or whether affiliate links pop up in answers.

That’s an example of how O’Connor sees the site helping users make their way through the clutter and chaos of the ton of data that is out there on the web, but that isn’t unified in a way that allows factual and objective apples-to-apples comparisons. “We go after a topic like ski resorts and organize the most important information in there and condense it,” he says. For example, “there are thousands of attributes of ski resorts but consumers make their decision based on six to ten of them. To gather that data – especially when the information is represented in different formats across hundreds of thousands of sites – in a consistent format is a difficult problem. That’s the challenge of the semantic web.”

And that’s why a lot of what FindtheBest does now is manually human-curated – a first step to organizing what’s on the web in a relevant way so that the semantic web is able to present that data in a meaningful and transparent way for computers and humans, the company says. Technology and tools add speed to the service, but today you still need people to figure out what the semantics are.

“We dub it expert source,” O’Connor says, with its researchers (not the crowds or crawlers ala Google Squared) sussing out what information is found on the web for various topics and then relating that information to questions users are likely to want answered (some 100 million Q& data-driven A pairs so far). For instance, accumulating the data on everything from credit cards’ annual fees to purchase and default APRs enables the creation of an app to help a consumer compare options among card offerings. In other words, it gives them the factual data they need to understand a question like “Which credit card’s total cost of ownership will be less?” While they can’t actually ask such a question directly today in natural language – instead, they’re asked what they’d like to compare—the search bar is being re-engineered to enable such capabilities.

The site also takes advantage of a couple of other ways to get and improve its information. That includes leveraging some government efforts to introduce transparency by posting more of federal or state databases on the web, though not necessarily in a consumable-friendly way that makes comparisons of, say, research spending very easy. “So we take that and put it into something consumers can use and supplement it,” he says. FindtheBest also wants knowledgable parties, including the companies represented in its categories and apps, to revise or claim their own listings, adding useful supplementary information or updating data as necessary, subject to site researchers’ approval.

When FindtheBest creates a comparison app, it first defines the relationship within a topic. – e.g. a group or class might be “summer camps” – and entities within that might include activities, ages, and locations. “We define all those attributes, then we go and accumulate the data,” says O’Connor. “Within “summer camps” we know how things relate to one another so we let them filter and sort based on those attributes. That’s easy.” But it’s also doing a lot of internal tagging to enable relating summer camps’ data to other data in its system or external on the web. “It’s more complicated to relate hundreds of millions of facts just within our system. But we’ll do everything from mash-ups to semantic relationships” there over the next few months.

FindtheBest sees opportunity to generate revenue through Google- or Bing-sponsored results, as well as to partner with publishers that would want content to be accompanied by interactive, updated and easily understood views into the data behind their articles. “That’s where we would hope to play,” says O’Connor. For instance, it could be the back-end comparison source for a newspaper’s guide to local area schools or hospitals or nursing homes.

In fact, FindtheBest sent the blog Grub Street its comparison app that incorporates details on NYC vs. SF — which includes data that San Franciscans spend almost three times as much on food and accommodations as New Yorkers — for use as part of its coverage of a food festival that pits NY Chefs against San Fran Chefs.

“We’re really going after the very highly structured data,” says O’Connor. “That’s where we see our expertise.”

SemTechBiz is Less Than 3 Weeks Away

The Semantic Tech & Business Conference (SemTechBiz) is coming to San Francisco on June 3-7! Join us for case studies, innovative panels, tutorials, and keynotes that will provide you with practical advice, hands-on guidance, and breakthrough approaches to solving business problems with semantic technology. Passes go up $200 at the door. Sign up now and save !