Using Label Information in a Genetic Programming Based Method for Acquiring Tag Tree Patterns with Vertex Labels and Wildcards

11 pages•Published: September 20, 2022

Shunsuke Yokoyama, Tetsuhiro Miyahara, Yusuke Suzuki, Tomoyuki Uchida and Tetsuji Kuboyama

Abstract

Machine learning and data mining from tree structured data are studied intensively. In this paper, as tree structured patterns we use tag tree patterns with vertex and edge labels and wildcards in order to represent label connecting relation of vertices and edges in tree structured data. We propose an evolutionary learning method based on Genetic Programming for acquiring characteristic tag tree patterns with vertex and edge labels and wildcards from positive and negative tree structured data. By using label information, that is, label connecting relation of positive examples, as inappropriate individuals we can exclude tag tree patterns that do not satisfy label connecting relation of positive examples. We report experimental results on our evolutionary learning method and show the effectiveness of using label connecting relation of positive examples.

Keyphrases: evolutionary learning method, genetic programming, tree structured patterns

In: Tokuro Matsuo (editor). Proceedings of 11th International Congress on Advanced Applied Informatics, vol 81, pages 418-428.

Links:	https://easychair.org/publications/paper/mMmH
	https://doi.org/10.29007/tfgn

BibTeX entry

@inproceedings{IIAIAAI2021-Winter:Using_Label_Information_Genetic,
  author    = {Shunsuke Yokoyama and Tetsuhiro Miyahara and Yusuke Suzuki and Tomoyuki Uchida and Tetsuji Kuboyama},
  title     = {Using Label Information in a Genetic Programming Based Method for Acquiring Tag Tree Patterns with Vertex Labels and Wildcards},
  booktitle = {Proceedings of 11th International Congress on Advanced Applied Informatics},
  editor    = {Tokuro Matsuo},
  series    = {EPiC Series in Computing},
  volume    = {81},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2398-7340},
  url       = {/publications/paper/mMmH},
  doi       = {10.29007/tfgn},
  pages     = {418-428},
  year      = {2022}}

Download PDF Open PDF in browser