All I'm starting to do a lit review on the subject of developing an ontology to perform semantic difference on classified terms. I've done a few google runs, and checked through citeseer, but since my specialty is machine learning, I'm a bit out of my element when it comes to ontologies/logic programming. The best way to find relevant papers would be to ask experts, but alas, I haven't networked with anybody who knows anything about this subject. So I turn to you, GameDev AI forum, for suggestions for good papers on this subject. To help you guys out, here's a little bit about the project that I'm working on: I'm working for a contractor that's been contracted out to help in the independent evaluation of several other contractors that are working on object recognition technologies. My project is to use an ontology to create a mechanical method to determine semantic difference between the actual object and the result returned by the object recognition technologies during the testing phase. For example, if we present one of the systems with an SUV and it classifies it as 2-door sedan, we want to be able to quantify the semantic difference between SUV and 2-door sedan. Any relevant literature would be welcome. Also, literature that would suggest that this isn't feasible is welcome too. If we're barking up the wrong tree, then we need to know that soon. And if helping somebody in need is not enough motivation, I also promise rate++ to any useful resources posted here.