Skip to main navigation Skip to search Skip to main content

Linguistic Decision Trees for Classification

  • Zhejiang University

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

Abstract

In this chapter, label semantics theory is applied to designing transparent data mining models. A label semantics based decision tree model is proposed where nodes are linguistic descriptions of variables and leaves are sets of appropriate labels. For each branch, instead of labeling it with a certain class, the probability of a particular class given this branch can be computed based on the given training dataset. This new model is referred to as a linguistic decision tree (LDT).

Original languageEnglish
Title of host publicationAdvanced Topics in Science and Technology in China
PublisherSpringer Science and Business Media Deutschland GmbH
Pages77-119
Number of pages43
DOIs
StatePublished - 2014

Publication series

NameAdvanced Topics in Science and Technology in China
ISSN (Print)1995-6819
ISSN (Electronic)1995-6827

Keywords

  • Focal Element
  • Information Granule
  • Label Semantic
  • Merging Algorithm
  • Threshold Probability

Fingerprint

Dive into the research topics of 'Linguistic Decision Trees for Classification'. Together they form a unique fingerprint.

Cite this