2018
DOI: 10.1109/tkde.2018.2808532
|View full text |Cite
|
Sign up to set email alerts
|

Unsupervised Coupled Metric Similarity for Non-IID Categorical Data

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
14
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 32 publications
(14 citation statements)
references
References 42 publications
0
14
0
Order By: Relevance
“…At present, there are few studies on qualitative analysis of nominal data, especially the data are Non-IID. Couple metric similarity (CMS) [ 36 ] is good for measuring the distance of Non-IID nominal data.…”
Section: Related Workmentioning
confidence: 99%
See 1 more Smart Citation
“…At present, there are few studies on qualitative analysis of nominal data, especially the data are Non-IID. Couple metric similarity (CMS) [ 36 ] is good for measuring the distance of Non-IID nominal data.…”
Section: Related Workmentioning
confidence: 99%
“…To facilitate the discussion in the remainder of this paper, CMS is reviewed briefly in this section. CMS measures the similarity of two objects by capturing both the intra- and inter-attribute coupling relations of objects, where the former characterizes the coupling similarity between the frequency distribution and the value of attribute and the latter aggregates attribute dependencies between different attribute values relationship by considering the intersection of the condition attribute values co-occurrence probability of the different characteristics [ 36 ].…”
Section: Preliminariesmentioning
confidence: 99%
“…By combining the intra-relationships and inter-relationships of attribute values, Coupled Object Similarity (COS) measure learned the similarity for categorical data [21]. Based on COS, Couple Metric Similarity (CMS) was proposed recently as a similarity metric [22]. However, in these works, only the pairwise relationship is considered, which neglected more possible complex relationships.…”
Section: Related Workmentioning
confidence: 99%
“…In reference to [22], the weight of < (a 1 , b 1 ), (a 2 , b 2 ) > can be derived from both the weights of < a 1 , a 2 > and < b 1 , b 2 >, which can be calculated bŷ…”
Section: A Heterogeneous Graph Constructionmentioning
confidence: 99%
See 1 more Smart Citation