The simility theory deals with the similarity of objects and conceptions, or something similar.
Everything is either the same or not. But not everything is either the same or different.
Things can only differ in quantity. In the quality they can also be similar.
The simility theory is based on geometry, metrics and topology, using the methods of multivariate statistics, cluster analysis, chaos mathematics and artificial intelligence.
Simility is a qualitative distance after a selected metric.
Its most important importance is again in the area of artificial intelligence. What is approximated there as "error function" or "fitness" actually means a similarity, that is, a distance after a certain metric.
The measure of similarity expresses itself through the greatest commonness, plus the similarities to the left and right of it.
The most demanding work in the modeling of self-learning systems is the description of a similarity context for the respective task. According to the above definition, this is about the metric quantification of qualities. To express this simply by means of an error value is often too short. Basically it is more about one or more distances. These provide the ability to distinguish what we call intelligence.