Arbeitspapier

K-expectiles clustering

K-means clustering is one of the most widely-used partitioning algorithm in cluster analysis due to its simplicity and computational efficiency, but it may not provide ideal clustering results when applying to data with non-spherically shaped clusters. By considering the asymmetrically weighted distance, We propose the K-expectile clustering and search the clusters via a greedy algorithm that minimizes the within cluster τ -variance. We provide algorithms based on two schemes: the fixed τ clustering, and the adaptive τ clustering. Validated by simulation results, our method has enhanced performance on data with asymmetric shaped clusters or clusters with a complicated structure. Applications of our method show that the fixed τ clustering can bring some flexibility on segmentation with a decent accuracy, while the adaptive τ clustering may yield better performance.

Language
Englisch

Bibliographic citation
Series: IRTG 1792 Discussion Paper ; No. 2021-003

Classification
Wirtschaft
Mathematical and Quantitative Methods: General
Subject
clustering
expectiles
asymmetric quadratic loss
image segmentation

Event
Geistige Schöpfung
(who)
Wang, Bingling
Li, Yingxing
Härdle, Wolfgang
Event
Veröffentlichung
(who)
Humboldt-Universität zu Berlin, International Research Training Group 1792 "High Dimensional Nonstationary Time Series"
(where)
Berlin
(when)
2021

Handle
Last update
10.03.2025, 11:43 AM CET

Data provider

This object is provided by:
ZBW - Deutsche Zentralbibliothek für Wirtschaftswissenschaften - Leibniz-Informationszentrum Wirtschaft. If you have any questions about the object, please contact the data provider.

Object type

  • Arbeitspapier

Associated

  • Wang, Bingling
  • Li, Yingxing
  • Härdle, Wolfgang
  • Humboldt-Universität zu Berlin, International Research Training Group 1792 "High Dimensional Nonstationary Time Series"

Time of origin

  • 2021

Other Objects (12)