@article{826,
  author = {Hamid Parvin, Zahra Rezaei, Sajad Parvin},
  title = {A Method to Construct Diverse Classifiers},
  journal = {Progress in Computing Applications},
  year = {2012},
  volume = {1},
  number = {1},
  doi = {},
  url = {http://www.dline.info/pca/fulltext/v1n1/4.pdf},
  abstract = {Usage of recognition systems has found many applications in almost all fields. Generally in design of combinational classifier systems, the more diverse the results of the classifiers, the more appropriate final result. However, Most of classification algorithms have obtained good performance for specific problems; they have not enough robustness for other problems. Combination of multiple classifiers can be considered as a general solution method for pattern recognition problems. It has been shown that combination of classifiers can usually operate better than single classifier provided that its components are independent or they have diverse outputs. It was shown that the necessary diversity of an ensemble can be achieved manipulation of data set features. We also propose a new method of creating this diversity. The ensemble created by proposed method may not always outperforms all classifiers existing in it, it is always possesses the diversity needed for creation of ensemble, and consequently it always outperforms the simple classifier.},
}