On the Problem Related to Reductive Attributes in the Incomplete Decision Tables

Authors

  • Janos Demetrovics Institute for Computer Science and Control (SZTAKI), Hungarian Academy of Sciences, Budapest, Hungary
  • Nguyen Long Giang Institute of Information Technology, Vietnam Academy of Science and Technology, Hanoi, Vietnam
  • Vu Duc Thi Information Technology Institute, Vietnam National University, Hanoi, Vietnam
  • Pham Viet Anh Hanoi University of Industry, Hanoi, Vietnam

DOI:

https://doi.org/10.55630/sjc.2022.16.24-38

Abstract

Attribute reduction is a key problem in the process of data mining and knowledge discovery. Up to now, many attribute reduction algorithms in incomplete decision tables have been proposed. However, the research results related to conditional attributes and reduct of incomplete decision tables are still limited. By relational database approach, this paper investigates some properties of conditional attributes and proposes an algorithm to determine all reductive attributes of consistent incomplete decision tables in polynomial time. The proposed algorithm is an effective tool to eliminate all redundant attributes in data pre-processing in order to improve the efficiency of data mining models.

Downloads

Published

2022-07-11

Issue

Section

Articles