본문 바로가기
  • Home

Classifier Capable of Handling Incomplete Data using Reparation Technique

  • Journal of Knowledge Information Technology and Systems
  • Abbr : JKITS
  • 2013, 8(2), pp.93-100
  • Publisher : Korea Knowledge Information Technology Society
  • Research Area : Interdisciplinary Studies > Interdisciplinary Research
  • Published : April 30, 2013

Lee Jong Chan 1 한기선 2

1청운대학교
2강동대학교

Accredited

ABSTRACT

Classification of the numerical data is a very important research topic in machine learning. But the incomplete data inwhich certain features are missing, is very common in real world applications. For solving this problem, a data reparationapproach base on the Fuzzy c-Means(FCM) clustering is used to estimate the incomplete data. This approach calculates thecentroid vectors of the clusters and then determined the membership probability, and repeat this process until the optimumsolution is found. Then a new method is proposed to classify the repaired data and it has an outstanding performance. Usually, classification problem can be separated into two phases: learning phase and classification phase. Many methodsdealing with incomplete data in classification problem have been proposed, but most of them only focus on the processing ofhandling incomplete data in learning phase. For the incomplete value appearing in the classification phase, almost all of thecurrent approaches can not work. So handling incomplete data in both learning and classification phase is important andnecessary to be applied for solving the real world problems.

Citation status

* References for papers published after 2023 are currently being built.