Limits...
Exploring sampling in the detection of multicategory EEG signals.

Siuly S, Kabir E, Wang H, Zhang Y - Comput Math Methods Med (2015)

Bottom Line: In the similar way, for the OS scheme, an OS set is obtained.Then eleven statistical features are extracted from the RS and OS set, separately.The experimental outcomes demonstrate that the RS scheme well represents the EEG signals and the k-NN with the RS is the optimum choice for detection of multicategory EEG signals.

View Article: PubMed Central - PubMed

Affiliation: Centre for Applied Informatics, College of Engineering and Science, Victoria University, P.O. Box 14428, Melbourne, VIC 8001, Australia.

ABSTRACT
The paper presents a structure based on samplings and machine leaning techniques for the detection of multicategory EEG signals where random sampling (RS) and optimal allocation sampling (OS) are explored. In the proposed framework, before using the RS and OS scheme, the entire EEG signals of each class are partitioned into several groups based on a particular time period. The RS and OS schemes are used in order to have representative observations from each group of each category of EEG data. Then all of the selected samples by the RS from the groups of each category are combined in a one set named RS set. In the similar way, for the OS scheme, an OS set is obtained. Then eleven statistical features are extracted from the RS and OS set, separately. Finally this study employs three well-known classifiers: k-nearest neighbor (k-NN), multinomial logistic regression with a ridge estimator (MLR), and support vector machine (SVM) to evaluate the performance for the RS and OS feature set. The experimental outcomes demonstrate that the RS scheme well represents the EEG signals and the k-NN with the RS is the optimum choice for detection of multicategory EEG signals.

Show MeSH

Related in: MedlinePlus

3D stacked area graph showing MAE for the k-NN, MLR, and SVM classifier under the RS and OS scheme.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4419228&req=5

fig8: 3D stacked area graph showing MAE for the k-NN, MLR, and SVM classifier under the RS and OS scheme.

Mentions: Figure 7 presents ROC areas for the k-NN, MLR, and SVM classifiers with the RS and OS scheme, separately for each of five classes and their overall ROC area as well. The area of the ROC curve is used as an index for evaluating classifier performance (e.g., lager area indicates better performance of the classifier). As can be seen in Figure 7, each of the three classifiers produces higher ROC area close to 1 with the use of the RS scheme for each class while they yield lower area with the use of the OS scheme. This figure validates the reliability of the use of the RS scheme compared with the OS scheme to get representative sample point from the EEG data. The shape of the MAE for each of the three classifiers under the RS and OS scheme is illustrated in Figure 8. It is noted that the lower MAE score indicates the higher performance of the scheme. We can see that the score of MAE is very low for the RS approach for each of the three classifiers. On the other hand, the OS approach yields very high score of MAE for each of the classifiers. In this figure, we also observe that the lowest MAE is produced by the k-NN approach among the three classifiers for the RS scheme. Thus we can argue strongly that the statistical features obtained from RS scheme are perfect representation of EEG signals and the k-NN classifier is the best choice for multicategory EEG signals detection.


Exploring sampling in the detection of multicategory EEG signals.

Siuly S, Kabir E, Wang H, Zhang Y - Comput Math Methods Med (2015)

3D stacked area graph showing MAE for the k-NN, MLR, and SVM classifier under the RS and OS scheme.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4419228&req=5

fig8: 3D stacked area graph showing MAE for the k-NN, MLR, and SVM classifier under the RS and OS scheme.
Mentions: Figure 7 presents ROC areas for the k-NN, MLR, and SVM classifiers with the RS and OS scheme, separately for each of five classes and their overall ROC area as well. The area of the ROC curve is used as an index for evaluating classifier performance (e.g., lager area indicates better performance of the classifier). As can be seen in Figure 7, each of the three classifiers produces higher ROC area close to 1 with the use of the RS scheme for each class while they yield lower area with the use of the OS scheme. This figure validates the reliability of the use of the RS scheme compared with the OS scheme to get representative sample point from the EEG data. The shape of the MAE for each of the three classifiers under the RS and OS scheme is illustrated in Figure 8. It is noted that the lower MAE score indicates the higher performance of the scheme. We can see that the score of MAE is very low for the RS approach for each of the three classifiers. On the other hand, the OS approach yields very high score of MAE for each of the classifiers. In this figure, we also observe that the lowest MAE is produced by the k-NN approach among the three classifiers for the RS scheme. Thus we can argue strongly that the statistical features obtained from RS scheme are perfect representation of EEG signals and the k-NN classifier is the best choice for multicategory EEG signals detection.

Bottom Line: In the similar way, for the OS scheme, an OS set is obtained.Then eleven statistical features are extracted from the RS and OS set, separately.The experimental outcomes demonstrate that the RS scheme well represents the EEG signals and the k-NN with the RS is the optimum choice for detection of multicategory EEG signals.

View Article: PubMed Central - PubMed

Affiliation: Centre for Applied Informatics, College of Engineering and Science, Victoria University, P.O. Box 14428, Melbourne, VIC 8001, Australia.

ABSTRACT
The paper presents a structure based on samplings and machine leaning techniques for the detection of multicategory EEG signals where random sampling (RS) and optimal allocation sampling (OS) are explored. In the proposed framework, before using the RS and OS scheme, the entire EEG signals of each class are partitioned into several groups based on a particular time period. The RS and OS schemes are used in order to have representative observations from each group of each category of EEG data. Then all of the selected samples by the RS from the groups of each category are combined in a one set named RS set. In the similar way, for the OS scheme, an OS set is obtained. Then eleven statistical features are extracted from the RS and OS set, separately. Finally this study employs three well-known classifiers: k-nearest neighbor (k-NN), multinomial logistic regression with a ridge estimator (MLR), and support vector machine (SVM) to evaluate the performance for the RS and OS feature set. The experimental outcomes demonstrate that the RS scheme well represents the EEG signals and the k-NN with the RS is the optimum choice for detection of multicategory EEG signals.

Show MeSH
Related in: MedlinePlus