Recently, I’ve blogged about:
- A Bayesian-ontological approach for identifying spam
- The limitations of fractal-dimension calculations in classifying in-situ observations of hydrometeors in clouds – and alluded to how knowledge-based methods may help to improve this situation
In both cases, there’s a case to be made for combining heuristic with knowledge-based approaches.
Although I did find “heuristics” and “knowledge” juxtaposed in Googling for “knowledge-based heuristics”, I believe the tightly coupled examples I’ve described above have some degree of novelty.
Further research is required 🙂
Ian Lumb's Blog 
“Kownledge”-based heuristics. I’m not sure if there is any other kind of heuristic. Ignorance-based knowledge?
Touche – even though this wasn’t the kind of novelty I was after 😉