New Method Based on Rough Set for Filling Missing Value


Cekik R., Telceken S.

6th World Conference on Soft Computing, California, United States Of America, 22 - 25 May 2016, vol.361, pp.41-48 identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 361
  • Doi Number: 10.1007/978-3-319-75408-6_4
  • City: California
  • Country: United States Of America
  • Page Numbers: pp.41-48
  • Keywords: Missing value, Rough set, Data mining
  • Anadolu University Affiliated: Yes

Abstract

The presence of missing value in a dataset can affect the performance of an analysis system such as classifier. To solve this problem many methods have been proposed in different studies using different theorems, analysis systems and methods such as Neural Network (NN), k-Nearest Neighbor (k-NN), closest fit etc. In this paper, we propose novel method based on RST for solving the problem of missing value that was lost (e.g., was erased). After dataset filling with proposed method, it has been observed improvement the performance of used analysis systems.