Dimension reduction in EEG data using Particle Swarm Optimization

Atyabi, Adham; Luerssen, Martin H.; Fitzgibbon, Sean P.; Powers, David

doi:10.1109/cec.2012.6256487

Cited by 13 publications

(17 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Atyabi et al [27] investigated the impact of evolutionary approaches such as Genetic Algorithm, Random Search, and PSO for feature and channel reduction among which the PSO-based feature reduction approach showed better overall generalizability. Atyabi et al [12] proposed 99% reduction through simultaneous reduction of feature and channel sets using a PSO-based approach with two layer swarm structure called PSO-DR. In the study, PSO-DR showed better overall performance in comparison to variety of approaches including 2 variations of Single Value Decomposition (i.e., SVD-based electrode reduction and SVD-based decomposition) and PSO-based electrode reduction.…”

Section: Pso Dimension Reduction (Pso-dr)mentioning

confidence: 99%

“…The choice of using sigmoid ELM as the classifier is made based on its fast learning capability and its time efficiency. Previous experiments indicated a lack of generalizability for the generated masks as they performed poorly with the unseen testing set [12]. The PSO-DR uses Linearly Decreasing Inertia Weight (LDIW) with w 1 = 0.2 and w 2 = 1 and fixed acceleration coefficients with c 1 = 0.5 and c 2 = 2.5 1 .…”

Section: Simultaneous Feature and Electrode Reduction With Pso-drmentioning

confidence: 99%

“…This issue has been resolved [12] by extracting three new masks representing the best performing masks on validation and testing sets (BestMask) and the mask containing the most commonly selected indexes from the groups of masks generated in stage 1 (ComMask). The potential of these new masks are assessed using a new 10 × 20 CV.…”

Section: Simultaneous Feature and Electrode Reduction With Pso-drmentioning

confidence: 99%

“…The potential of these new masks are assessed using a new 10 × 20 CV. Figure 3 provides a graphical representation of the procedures performed in the two stages in [12].…”

Section: Simultaneous Feature and Electrode Reduction With Pso-drmentioning

confidence: 99%

See 3 more Smart Citations

Reducing training requirements through evolutionary based dimension reduction and subject transfer

et al. 2017

Self Cite

View full text Add to dashboard Cite

Training Brain Computer Interface (BCI) systems to understand the intention of a subject through Electroencephalogram (EEG) data currently requires multiple training sessions with a subject in order to develop the necessary expertise to distinguish signals for different tasks. Conventionally the task of training the subject is done by introducing a training and calibration stage during which some feedback is presented to the subject. This training session can take several hours which is not appropriate for on-line EEG-based BCI systems. An alternative approach is to use previous recording sessions of the same person or some other subjects that performed the same tasks (subject transfer) for training the classifiers. The main aim of this study is to generate a methodology that allows the use of data from other subjects while reducing the dimensions of the data. The study investigates several possibilities for reducing the necessary training and calibration period in subjects and the classifiers and addresses the impact of i) evolutionary subject transfer and ii) adapting previously trained methods (retraining) using other subjects data. Our results suggest reduction * Corresponding author to 40% of target subject data is sufficient for training the classifier. Our results also indicate the superiority of the approaches that incorporated evolutionary subject transfer and highlights the feasibility of adapting a system trained on other subjects.

show abstract

Section: Pso Dimension Reduction (Pso-dr)mentioning

confidence: 99%

Section: Simultaneous Feature and Electrode Reduction With Pso-drmentioning

confidence: 99%

Section: Simultaneous Feature and Electrode Reduction With Pso-drmentioning

confidence: 99%

“…The potential of these new masks are assessed using a new 10 × 20 CV. Figure 3 provides a graphical representation of the procedures performed in the two stages in [12].…”

Section: Simultaneous Feature and Electrode Reduction With Pso-drmentioning

confidence: 99%

See 2 more Smart Citations

Reducing training requirements through evolutionary based dimension reduction and subject transfer

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…As far as the authors' knowledge is concerned, although PSO has been employed to address various issues in BCI applications such as feature selection [21][22][23][24], source localization [25,26], change point detection [27] and adaptive signal filtering [28,29], it has only been employed as static classifier, in which PSO has mostly been utilized as training algorithm for the neural classifier. PSO-based RBFNN [30] and PSO-based recurrent NN [31] are examples of these hybrid dynamic classifiers.…”

Section: Related Workmentioning

confidence: 99%

An incremental framework for classification of EEG signals using quantum particle swarm optimization

Hassani

Lee

2014

2014 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applicati

View full text Add to dashboard Cite

Classification of electroencephalographic (EEG) signals is a sophisticated task that determines the accuracy of thought pattern recognition performed by computer-brain interface (BCI) which, in turn, determines the degree of naturalness of the interaction provided by that system. However, classifying the EEG signals is not a trivial task due to their nonstationary characteristics. In this paper, we introduce and utilize incremental quantum particle swarm optimization (IQPSO) algorithm for incremental classification of EEG data stream. IQPSO builds the classification model as a set of explicit rules which benefits from semantic symbolic knowledge representation and enhanced comprehensibility. We compared the performance of IQPSO against ten other classifiers on two EEG datasets. The results suggest that IQPSO outperforms other classifiers in terms of classification accuracy, precision and recall.

show abstract