Summarization of information systems based on rough set theory

Yen Liang Chen, Fang Chi Chi

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

In the rough set theory proposed by Pawlak, the concept of reduct is very important. The reduct is the minimum attribute set that preserves the partition of the universe. A great deal of research in the past has attempted to reduce the representation of the original table. The advantage of using a reduced representation table is that it can summarize the original table so that it retains the original knowledge without distortion. However, using reduct to summarize tables may encounter the problem of the table still being too large, so users will be overwhelmed by too much information. To solve this problem, this article considers how to further reduce the size of the table without causing too much distortion to the original knowledge. Therefore, we set an upper limit for information distortion, which represents the maximum degree of information distortion we allow. Under this upper limit of distortion, we seek to find the summary table with the highest compression. This paper proposes two algorithms. The first is to find all summary tables that satisfy the maximum distortion constraint, while the second is to further select the summary table with the greatest degree of compression from these tables.

Original languageEnglish
Pages (from-to)1001-1015
Number of pages15
JournalJournal of Intelligent and Fuzzy Systems
Volume40
Issue number1
DOIs
StatePublished - 2021

Keywords

  • Attribute reduction
  • Information system
  • Reduct
  • Rough set
  • Summarization

Fingerprint

Dive into the research topics of 'Summarization of information systems based on rough set theory'. Together they form a unique fingerprint.

Cite this