Applying Rule Post-Pruning to Condition-based Tree for Generating Concise Rules - a Simple, but Effective Method

The big problem with associative classifiers is that they contain a large number of rules that make them difficult to interpret. To deal with this challenge, The authors in work [8] have represented this set ofrules as a tree model, namely condition-based tree (CBT), and propose an algorithm to conv...

Full description

Saved in:
Bibliographic Details
Published in2022 RIVF International Conference on Computing and Communication Technologies (RIVF) pp. 743 - 748
Main Authors Suong, Tieu P. M., Cuong, Nguyen D.
Format Conference Proceeding
LanguageEnglish
Published IEEE 20.12.2022
Subjects
association rule
associative classifier
Classification algorithms
Communications technology
decision tree
rule pruning
Online AccessGet full text

Cover

More Information
Summary:The big problem with associative classifiers is that they contain a large number of rules that make them difficult to interpret. To deal with this challenge, The authors in work [8] have represented this set ofrules as a tree model, namely condition-based tree (CBT), and propose an algorithm to convert this tree into a more concise rule set. In this paper, the authors inherit the idea of [8] and perform fine-turning on CBT by applying Rule Post-Pruning before converting this tree into rule set. Experimental results show that the proposed classifier has generated a small number of association rules and these rules give high accuracy compared to well-known associative classifiers.
DOI:10.1109/RIVF55975.2022.10013884