Missing Values Imputation Using Fuzzy C Means Based on Correlation of Variable
Missing values is one of the problems in real-world data and an unavoidable one. It should be handled carefully in a pre-processing technique before being processed in a data mining technique. This paper proposes an imputation technique of Fuzzy C Mean (FCM) with the improved version. The aim is to...
| Main Authors: | Mausor, F.H., Jaafar, J., Taib, S.M. |
|---|---|
| Format: | Conference or Workshop Item |
| Institution: | Universiti Teknologi Petronas |
| Record Id / ISBN-0: | utp-eprints.29867 / |
| Published: |
Institute of Electrical and Electronics Engineers Inc.
2020
|
| Online Access: |
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85097526325&doi=10.1109%2fICCI51257.2020.9247675&partnerID=40&md5=a41e37dcac4c9b74c59ad899dfd60d1c http://eprints.utp.edu.my/29867/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: |
Missing values is one of the problems in real-world data and an unavoidable one. It should be handled carefully in a pre-processing technique before being processed in a data mining technique. This paper proposes an imputation technique of Fuzzy C Mean (FCM) with the improved version. The aim is to reduce errors and increase the accuracy of the processing technique. In this paper, the correlation technique was applied before the process of FCM to choose the variables with a certain criterion to be processed in FCM imputation. The result shows that the proposed technique outperforms the conventional technique and useful to overcome the disadvantages of the FCM technique. © 2020 IEEE. |
|---|