Comparison of partial orders clustering techniques

In this paper, we compare three approaches of clustering partial ordered subsets of a set of items. First approach was k-medoids clustering algorithm with distance function based on Levenshtein distance. The second approach was k-means algorithm with cosine distance as distance function after vector...

Full description

Bibliographic Details
Main Author:	A. Raskin
Format:	Article
Language:	English
Published:	Ivannikov Institute for System Programming of the Russian Academy of Sciences 2018-10-01
Series:	Труды Института системного программирования РАН
Subjects:	расстояние левенштейна частично упорядоченные множества кластеризация меры близости коэффициент корреляции кендалла
Online Access:	https://ispranproceedings.elpub.ru/jour/article/view/829

id	doaj-f3b5a70d3c344bb488302682380df929
record_format	Article
spelling	doaj-f3b5a70d3c344bb488302682380df9292020-11-25T02:16:36Zeng Ivannikov Institute for System Programming of the Russian Academy of SciencesТруды Института системного программирования РАН2079-81562220-64262018-10-01264919810.15514/ISPRAS-2014-26(4)-7829Comparison of partial orders clustering techniquesA. Raskin0Национальный исследовательский ядерный университет «МИФИ»In this paper, we compare three approaches of clustering partial ordered subsets of a set of items. First approach was k-medoids clustering algorithm with distance function based on Levenshtein distance. The second approach was k-means algorithm with cosine distance as distance function after vectorization of partial orders. And the third one was k-medoids algorithm with Kendall's tau as a distance function. We use Adjusted Rand Index as a measure of quality of clustering and find out that clustering with all three methods get stable results when variance of number of items ranked is high. Vectorization of partial orders get best results if number of items ranked is low.https://ispranproceedings.elpub.ru/jour/article/view/829расстояние левенштейначастично упорядоченные множествакластеризациямеры близостикоэффициент корреляции кендалла
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	A. Raskin
spellingShingle	A. Raskin Comparison of partial orders clustering techniques Труды Института системного программирования РАН расстояние левенштейна частично упорядоченные множества кластеризация меры близости коэффициент корреляции кендалла
author_facet	A. Raskin
author_sort	A. Raskin
title	Comparison of partial orders clustering techniques
title_short	Comparison of partial orders clustering techniques
title_full	Comparison of partial orders clustering techniques
title_fullStr	Comparison of partial orders clustering techniques
title_full_unstemmed	Comparison of partial orders clustering techniques
title_sort	comparison of partial orders clustering techniques
publisher	Ivannikov Institute for System Programming of the Russian Academy of Sciences
series	Труды Института системного программирования РАН
issn	2079-8156 2220-6426
publishDate	2018-10-01
description	In this paper, we compare three approaches of clustering partial ordered subsets of a set of items. First approach was k-medoids clustering algorithm with distance function based on Levenshtein distance. The second approach was k-means algorithm with cosine distance as distance function after vectorization of partial orders. And the third one was k-medoids algorithm with Kendall's tau as a distance function. We use Adjusted Rand Index as a measure of quality of clustering and find out that clustering with all three methods get stable results when variance of number of items ranked is high. Vectorization of partial orders get best results if number of items ranked is low.
topic	расстояние левенштейна частично упорядоченные множества кластеризация меры близости коэффициент корреляции кендалла
url	https://ispranproceedings.elpub.ru/jour/article/view/829
work_keys_str_mv	AT araskin comparisonofpartialordersclusteringtechniques
_version_	1724890213083250688

Comparison of partial orders clustering techniques

Similar Items