2021
DOI: 10.1016/j.knosys.2021.106749
|View full text |Cite
|
Sign up to set email alerts
|

Semi-supervised classification on data streams with recurring concept drift and concept evolution

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
16
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
7
2
1

Relationship

0
10

Authors

Journals

citations
Cited by 47 publications
(16 citation statements)
references
References 42 publications
0
16
0
Order By: Relevance
“…Many methods can then deploy this phase (e.g., K-means [7] and Kolmogorov-Smirnov (KS) statistics test [9]). The choice may depend on the learning setting, i.e., unsupervised, supervised, [6], or semisupervised [10]. The hypothesis test calculator analyzes and questions the accuracy of the report sent by the testing calculator.…”
Section: A Concept-drift Detectionmentioning
confidence: 99%
“…Many methods can then deploy this phase (e.g., K-means [7] and Kolmogorov-Smirnov (KS) statistics test [9]). The choice may depend on the learning setting, i.e., unsupervised, supervised, [6], or semisupervised [10]. The hypothesis test calculator analyzes and questions the accuracy of the report sent by the testing calculator.…”
Section: A Concept-drift Detectionmentioning
confidence: 99%
“…These weights indicate the best corresponding model for the current concept [29]. Ensemblebased approaches dealing with concept drift are categorized into homogeneous and heterogeneous.…”
Section: Literature Reviewmentioning
confidence: 99%
“…Hence, when a concept drift happens, the new rumour subgraphs would no longer be generated according to the old stationary distribution. Then, the change in the distributions can be identified using the CUSUM test, see [47], which is based on the Central Limit Theorem for multivariate vector streams.…”
Section: Dealing With Concept Driftmentioning
confidence: 99%