الفهرس | Only 14 pages are availabe for public view |
Abstract The Neutrosophic set was introduced by Smarandache in 1995 as an extension to the fuzzy set. It has laid the foundation for a whole family of new mathematical theories generalizing both their classical and fuzzy counterparts, such as the Neutrosophic set theory, the Neutrosophic probability, the neutrosophic statistics and the neutrosophic logic. In this thesis, we generalize fuzzy systems to their neutrosophic counterparts. Because of the ability of neutrosophic sets to deal with imprecision, fuzzy, vague, incomplete, and inconsistency data, the main interest was to introduce better algorithms for clas- si cation and clustering problems. First, we have built a Neutrosophic Rule-based System for classi cation.Then we built a hybrid system between genetic algorithm and neutrosophic logic to optimize the performance of the rule-based system.Then, for clustering, we have developed a New Neutrosophic C-Means algorithm which gen- eralizes the Fuzzy C-Means algorithm and performs better |