In the introductory Section, we described a specific learning problem where positive and negative examples are given. In practice different variations of similar problems occur.
Positive and Negative Examples
Let the name of the background ontology be O. The goal in this learning problem is to find an OWL class expression C such that all/many positive examples are instances of C w.r.t. O and none/few negative examples are instances of C w.r.t. O. As explained previously, C should be learned such that it generalises to unseen individuals and is readable.
This learning problem is similar to the one before, but without negative examples. In this case, it is desirable to find a class expression which closely fits the positive examples while still generalising sufficiently well. For instance, you usually do not want to have
owl:Thing as a solution for this problem, but neither do you want to have an enumeration of all examples.
In class learning, you are given an existing class A within your ontology O and want to describe it. It is similar to the previous problem in that you can use the instances of the class as positive examples. However, there are some differences, e.g. you do not want to have A itself as a proposed solution of the problem, and since this is an ontology engineering task, the focus on short and readable class expressions is stronger than for the two problems mentioned before. The learner can also take advantage of existing knowledge about the class to describe.