Machine Learning Classifiers for Clumps in Binary Sequences

H. Lim, A.L. Goel (USA), and M. Shin (Korea)


Clumps, Machine learning, Classification, Radial Basis Functions


In this paper we develop a set of classification models for clumps that are subsequences of 1's in a binary sequence of 0's and 1's. The models we consider use Gaussian basis functions and the newly developed SG algorithm [1] for their design. We present models for two cases based on how training and test data are created. Then we compare their performance (classification error on training and test data) with the reported results [2] from C4.5 and CART algorithms and the MLP network. We conclude by observing that the new models outperform the other classifiers for the cases considered here.

Important Links:

Go Back