Rule induction for screening Thalassemia using machine learning techniques: C5.0 and CART

Patcharaporn Paokanta, Michele Ceccarelli, Napat Harnpornchai, Nopasit Chakpitak, Somdet Srichairatanakool

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Rule induction has played an important role in implementing a medical expert system in the past decade, especially the Thalassemia Expert System. Due to the fact that Thalassemia indicators used in diagnosising types of Thalassemia are very complex, the induction rules C5.0 and Classification and Regression Tree (CART) will be used to elicit new information about Thalassemia. The results obtained from using both algorithms show the different rules separating types of this disease. In the future, these results will be used to develop the Thalassemia Expert System and these results will be compared to find a suitable algorithm. Other algorithms will also be considered.

Original languageEnglish
Pages (from-to)301-306
Number of pages6
JournalICIC Express Letters
Volume6
Issue number2
Publication statusPublished - 1 Feb 2012
Externally publishedYes

    Fingerprint

Keywords

  • C5.0
  • Classification and regression tree (CART)
  • Decision tree
  • Rule induction
  • Thalassemia
  • Variable importance

ASJC Scopus subject areas

  • Computer Science(all)
  • Control and Systems Engineering

Cite this

Paokanta, P., Ceccarelli, M., Harnpornchai, N., Chakpitak, N., & Srichairatanakool, S. (2012). Rule induction for screening Thalassemia using machine learning techniques: C5.0 and CART. ICIC Express Letters, 6(2), 301-306.