Discovering accurate and common characteristic rules from large tables

Yu Chin Liu, Ping Yu Hsu

Research output: Contribution to journalArticlepeer-review

Abstract

With the wide installation of eBusiness and database softwaze in enterprises, mountains of data are accumulating in the form of relational tables. Discovering valuable information from the sea of data is of interest to researchers and managers worldwide. In this paper, an algorithm is proposed to find characteristics from a large database table. It can be applied to fmd chazacteristics of customers in a particular segments or the characteristics of patients,., etc. In contrast to traditional data generalization or induction methods, the proposed new method, named Char, does not need a concept tree in advance and can generate a manual set of characteristic rules that are precise enough to describe the main characteristics of the data. The simulation results show that the characteristic rules found by Char are efficient as well as consistent regardless of the number of records and of attributes in the dataset.

Original languageEnglish
Pages (from-to)31-40
Number of pages10
JournalIntelligent Automation and Soft Computing
Volume17
Issue number1
DOIs
StatePublished - Jan 2011

Keywords

  • Characteristics rules
  • Data mining
  • Entropy
  • Information loss
  • Redundancy

Fingerprint

Dive into the research topics of 'Discovering accurate and common characteristic rules from large tables'. Together they form a unique fingerprint.

Cite this