Learning of Glycan Motifs Using Genetic Programming and Various Fitness Functions
Tetsuhiro Miyahara, Tetsuji Kuboyama · Journal of Advanced Computational Intelligence and Intelligent Informatics · 2014
We apply a genetic programming approach to learning of glycan motifs by using tag tree patterns and various fitness functions. Tag tree patterns obtained from some glycan data show characteristic tree structures. We examine the effects of using various fitness functions on GP processes and obtained glycan motifs. We also show that our method is applicable to tree structured data other than glycan data.