2009
DOI: 10.1016/j.compbiomed.2009.01.007
|View full text |Cite
|
Sign up to set email alerts
|

Using conditional FCM to mine event-related brain dynamics

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
8
0

Year Published

2010
2010
2018
2018

Publication Types

Select...
4
2

Relationship

1
5

Authors

Journals

citations
Cited by 8 publications
(8 citation statements)
references
References 31 publications
0
8
0
Order By: Relevance
“…To prevent this problem, a semi-supervised learning would be preferred to un-supervised learning. Zigkolis and Laskaris (2009) developed a conditional FCM (CFCM) algorithm to tackle this problem. In this algorithm, a dissimilarity measure (expressed in terms of the euclidean distance) was minimized for the data points in the pre-defined clusters.…”
Section: Literature Surveymentioning
confidence: 99%
“…To prevent this problem, a semi-supervised learning would be preferred to un-supervised learning. Zigkolis and Laskaris (2009) developed a conditional FCM (CFCM) algorithm to tackle this problem. In this algorithm, a dissimilarity measure (expressed in terms of the euclidean distance) was minimized for the data points in the pre-defined clusters.…”
Section: Literature Surveymentioning
confidence: 99%
“…Our point of departure was previously published work on the related problem of single-trial analysis (Laskaris and Ioannides, 2002;Laskaris et al, 2004Laskaris et al, , 2008Zigkolis and Laskaris, 2009), in which evoked brain responses were detected organised, and visualised by exploiting the recent advances in non-linear dimensionality reduction. Specifically, we employ isometric feature mapping (ISOMAP), which is known to reveal the intrinsic data variation and is therefore expected to be insensitive to random variations due to noise.…”
Section: Introductionmentioning
confidence: 99%
“…Our approach constitutes an extension of a well-established framework for mining single-trial brain dynamics from multichannel encephalographic signals [72]- [76]. It was originally conceived as a toolbox of unsupervised learning procedures, with VQ playing an instrumental role in deriving faithful summaries of response variability and further enabling semantically rich visualizations.…”
Section: Preliminariesmentioning
confidence: 99%
“…Based on (one particular instantiation of) the test-set of single-trials recorded at the particular sensor, and using 𝑑 𝑒 = 200 and 𝜏 = 1 for reconstructing the response dynamics, the initial codebook was derived and presented as shown in Figure 10a. In this particular plot, the code-vectors have been ranked based on a graph-seriation procedure [76], [82], [83] (in order to spot the underlying similarities). By means of classic MDS, this codebook has been embedded in a 2D space in the form of a scatter plot that reflects the redundancies in codebook design as densely populated neighborhoods (Figure 10b).…”
Section: A Detailed Demonstrationmentioning
confidence: 99%
See 1 more Smart Citation