The Soft Computing paradigm refers to a collection of computational techniques in computer science, artificial intelligence, machine learning, and many other applied and engineering areas where one tries to study, model, and analyse very complex phenomena, those for which more precise scientific tools of the past were incapable of giving low cost, analytic, and complete solution.

The main idea of soft computing methods is to utilise and, possibly, take advantage of impreciseness, vagueness and approximate description that is intrinsic to real life problems. Also, soft computing solutions are built with tolerance for incomplete representation, partial truth and imprecise reasoning schemes. That comes from observation that more complex systems related to biology and medicine, humanities, management sciences, and similar fields remained outside of the main territory of successful applications of precise mathematical, and analytical methods.

Soft computing area is to large extent similar to what is called “nouvelle AI”. It is frequently associated with approaches based on fuzzy sets and fuzzy logic, but other areas such as rough sets, artificial neural networks, Bayesian reasoning, evolutionary computations, chaos theory, non-monotonic reasoning may also be considered within the scope of soft computing. Important feature of soft computing approach is the use of various techniques in such the way that they complement each other and in this way make it possible to achieve results that are unreachable by any single method.