Systems, Man, and Cybernetics

 

Home

• [P955] Splice Site Detection in DNA Sequences Using a Fast Classification Algorithm
  Author(s): Xiaoou Li, Wen Wen Yu
  
  

  Abstract
  Support Vector Machines (SVMs) are well known excellent algorithms for classification problem. The principal disadvantage of SVMs is due to their excessive training time for large data set, such as DNA sequences. This paper presents a novel SVMs classification method, which reduces significantly the input data set using Bayesian technique. Meanwhile, classification accuracy is not worse. The algorithm has been successfully applied on splice site detection in DNA sequences. For these huge data sets, experimental results show that the accuracy obtained by the proposed algorithm is comparable (98.2) with the other SVM classifications, such as SMO (98.4%), LibSVM (98.4%), and Simple SVM (97.6%). However, convergence speed of the proposed algorithm is faster than them.

  
   

Full Text (PDF)