The second approach uses the pattern library in an inverse fashion, meaning that the library encodes only the positive patterns, which are marked with green plus signs in the following diagram. When an observed behavior (the blue circle) cannot be matched against the library, it is considered anomalous:

This approach requires us to model only what we have seen in the past, that is, normal patterns. If we return to the doctor example, the main reason that we visited the doctor in the first place was because we did not feel well. Our perceived state of feelings (for example, a headache and sore skin) did not match our usual feelings, and therefore, we decided to seek a doctor. We don't know which disease caused this state, nor do we know the treatment, but we were able to observe that it doesn't match the usual state.

A major advantage of this approach is that it does not require us to say anything about abnormal patterns; hence, it is appropriate for modeling known unknowns and unknown unknowns. On the other hand, it does not tell us exactly what is wrong.