Fuzzy Clustering Methods for Incomplete Spherical Data

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00348
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Soft computing
• Research Institution: Shibaura Institute of Technology
• Principal Investigator: Kanzawa Yuchi 芝浦工業大学, 工学部, 教授 (00298176)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: ファジィクラスタリング / 球面データ / 不完全データ

Outline of Final Research Achievements

Clustering is applicable in many fields as an important data analysis tool. The objective of this research is constructing fuzzy clustering methods for incomplete spherical data, which will make fuzzy clustering methods for spherical data applicable to practical situations.
In 2015 fiscal year, we arranged fuzzy clustering methods for complete spherical data, and evaluated their performance for incomplete data by whole data strategy. In 2016 fiscal year, we compared the proposed methods with the conventional methods in terms of clustering accuracy using artificial data sets. In 2017 fiscal year, we constructed imputation methods for incomplete spherical data, which is corresponding with the complete information maximum likelihood estimation for linear statistics.

Free Research Field

ファジィクラスタリング