Not to go all metaphysical on you, but what is the ontology of good? At South by Southwest this week, DailyFeats will present its new social platform that plays off the idea: It lets people check in and be rewarded for their positive actions, via  the web, SMS, email, Foursquare and Google Talk.

In private beta the last few months (and don’t you wish you knew about it before your new year’s resolutions were just memories), co-founder and CTO Vinay Gidwaney says a conscious decision was made to classify the good, in order to assess into what categories to group a positive action and how various behaviors add up to rewards like badges. A veteran of the MIT Media Lab that launched the ConceptNet common-sense reasoning toolkit, Gidwaney explains that DailyFeats can parse out the feat to be recognized from an SMS message such as trudged a garbage bag full of old clothes to Good Will today to respond appropriately with Awesome – you did give it away.

There are about 200 feats so far represented, each of which incorporates detailed semantic understanding with respect to words that relate to them. “There’s a corpus for every feat developed and connections between feats are established,” he says. It’s rung up more than 100,000 check-ins so far, which is adding to the body of knowledge around people’s behaviors – not only the data point of the feat they check in itself, but when and in what order they add feats, how routine they are, and whether those activities are more reflective or declarative – that will help it drive ‘good’ behavior as the particular individual understands it.

“We think about how to get them to floss more, get people to eat better and how to use the web to activate that — how that happens from an ontological point of view, an understanding of people’s personalities and behavior, and connections with other data points for us to suggest and nudge them to do more good,” Gidwaney says. “What we try to do is make it available and accessible for everyone to do good across a wide breath of things and for that we need to understand them and to semantically understand them.” Recommendations on good actions you can take that conform to who you are can be delivered “because we know more about your lifestyle, how you classify and think about good, so we can recommend new feats for you, too.”

Data points that cumulatively indicate that an individual, for instance, is interested in saving money – clipping coupons, packing a lunch, and so on – lead to a get rich slowly badge reward. “We put a lot of thought into the ontology of badges and to get the get rich slowly one, you have to indicate you think consistently about saving money,” he says. Once you check in a certain number of feats that the service has related to the practice of saving money, you get the badge – and as long as you stay engaged in those activities, you keep it.

Part of the challenge for the company, when it thinks about semantic technology, is understanding the connections not just of apples to fruits and veggies but of emotions and aspirations to doing good. “We’ve a fairly good meta-understanding of the English language but as a community we have little understanding of the soft knowledge around emotion, aspiration and doing good and actively building that out,”  Gidwaney says. “That’s a wealth of knowledge we turn daily into product improvements and enhancements.”

That also might feed into the service’s business model as it looks to building positive behavior graphs that sponsors can tap into – in aggregate, not individually, of course – to connect their brand purpose with users’ personal purposes. DailyFeats provides partners with a dashboard-view of insights around things like consumer intent, how they plan on changing their lives and how brands can play role.

The site already has partner sponsors of feats such  like Monster.com – its business is being a job search engine but whose brand mission is to put people in the right environment to achieve their potential. By sponsoring a feat like !doitnow, with all its implications for reaching your goals and realizing your potential, Monster hopes to make the connection in users’ heads to think about it as a resource when changing careers. “A lot of companies are still a bit confused around social media,” Gidwaney says. “They can set up a Facebook page and be followed on Twitter, but the involvement of consumers with brands there isn’t huge. Part of what is missing is purpose, what impact that brand has on something I believe in. That’s where DailyFeats can be different.” In sponsoring feats, companies also contribute to the points users rack up that can be traded in the real-world as currency – exchange them for complementary museum trips, or discounts on various services.

Speaking of Facebook, Gidwaney sees opportunities to coordinate what it’s doing with things like the OpenGraph protocol and the like button, who people’s friends are there and other data, so that when they’re recommended a feat it might have some connection to what other members of their social community do or like. “We will tap into that as much we can to make the experience much better for our members,” he says.

SemTechBiz is Less Than 2 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 !