基于AdaBoost的改進模糊分類規(guī)則集成學習
Advance Ensemble Learning of Fuzzy Classification Rules Based on AdaBoost
-
摘要: 基于集成學習提出了一種新的模糊分類規(guī)則的產生算法。將分類規(guī)則的前件、后件模糊化,在自適應提升(Adaptive Boosting,AdaBoost)算法的迭代中,調整訓練實例的分布,利用遺傳算法產生模糊分類規(guī)則。并在規(guī)則學習的適應度函數中引入訓練實例的分布,使得模糊分類規(guī)則在產生階段就考慮相互之間的協作,產生具有互補性的分類規(guī)則集。從而改善了模糊分類規(guī)則的整體識別能力,提高了分類識別精度。Abstract: A new learning algorithm of fuzzy classification rules is presented based on ensemble learning algorithm. By tuning the distribution of training instances during each AdaBoost iterative training, the classification rules with fuzzy antecedent and consequent are produced with genetic algorithm. The distribution of training instances participate in computing of the fitness function and the collaboration of rules which are complementary is taken into account during rules producing, so that the classification error rate is reduced and performance of the classification based on the fuzzy rules is improved.
-
Cordon O, del Jesus M J. Genetic learning of fuzzy classification systems cooperating with fuzzy reasoning methods [J].International Journal of Intelligent Systems.1998, 13(10/11):1025-3.0.CO;2-N' target='_blank'>[2]Merz C J. Using correspondence analysis to combine classifiers[J]. Machine Learning, 1999 36(1/2): 33 - 58.[3]Freund Y, Schapire R E. Experiments with a new boosting algorithm[C]. Proc. of the Thirteenth International Conference on Machine Learning, Morgan Kaufmann, 1996: 148- 156.[4]Gonzalez A, Herrera F. Multi-stage genetic fuzzy systems based on the iterative rule learning approach[J].Mathware Soft Computing.1997, 4:233-[5]Schapire R E. Theoretical views of boosting[C]. In Proc. 4th European Conference on Computational Learning Theory, 1999:1-10.[6]Freund Y. Boosting a weak learning algorithm by majority[J].Information and Computation, 1995, 2(121): 256 - 285.[7]Ishibuchi H, Nozaki K, Yamamoto N, et al.. Selecting fuzzy if-then rules for classification problems using genetic algorithms[J].IEEE Trans. on Fuzzy Systems.1995, 3(2):260- -
計量
- 文章訪問數: 2234
- HTML全文瀏覽量: 111
- PDF下載量: 684
- 被引次數: 0
下載:
