Loading...

A novel fuzzy genetic annealing classification approach

Baran Pouyan, M ; Sharif University of Technology

698 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/EMS.2009.32
  3. Abstract:
  4. In this paper, a novel classification approach is presented. This approach uses fuzzy if-then rules for classification task and employs a hybrid optimization method to improve the accuracy and comprehensibility of obtained outcome. The mentioned optimization method has been formulated by simulated annealing and genetic algorithm. In fact, the genetic operators have been used as perturb functions at the core of simulated annealing heuristic. Results of proposed approach have been compared with several well-known methods such as Naïve Bayes, Support Vector Machine, Decision Tree, k-NN, and GBML, and show that our method performs the classification task as well as other famous algorithms. © 2009 IEEE
  5. Keywords:
  6. Classification ; Genetic algorithm ; Classification approach ; Classification tasks ; Fuzzy if-then rules ; Genetic algorithm simulated annealing ; Genetic operators ; Hybrid optimization method ; Optimization method ; Annealing ; Computer simulation ; Decision trees ; Fuzzy rules ; Genetic algorithms ; Magnetic susceptibility ; Mathematical operators ; Simulated annealing
  7. Source: EMS 2009 - UKSim 3rd European Modelling Symposium on Computer Modelling and Simulation, 25 November 2009 through 27 November 2009, Athens ; 2009 , Pages 87-91 ; 9780769538860 (ISBN)
  8. URL: http://ieeexplore.ieee.org/document/5358830