Clustering on the cylinder, sphere and torus-Statistical machine learning approaches

Sen Gupta A., Roy M.

1st International Conference on Computational Science and Data Analytics, COMDATA 2021, Incorporating the 1st South-East Asia Workshop on Computational Physics and Data Analytics, CPDAS 2021, Hybrid, Kuala Lumpur, Malezya, 21 - 24 Kasım 2021, cilt.2756 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 2756
  • Doi Numarası: 10.1063/5.0141859
  • Basıldığı Şehir: Hybrid, Kuala Lumpur
  • Basıldığı Ülke: Malezya
  • Orta Doğu Teknik Üniversitesi Adresli: Hayır

Özet

With the advent of science and technology, manifold data are playing novel and important roles in emerging real-life challenging problems. These Big Data pose new challenges in the realm of Statistical Machine Learning, e.g. in cluster analysis or unsupervised learning (in the machine learning parlour) of directional data. In this paper, a brief overview of the some of recent clustering methods enhanced for cylindrical data, spherical data and toroidal data are presented. A new distance-based method for clustering data on the torus is also indicated.