کنفرانس مهندسی برق ایران

صفحه اصلی / سی و یکمین کنفرانس بین المللی مهندسی برق

Kernel-Based Embedded Feature Selection for Motor Imagery Based BCI

نویسندگان :

Mehdi Kamandar¹

1- دانشگاه تحصیلات تکمیلی صنعتی و فناوری پیشرفته

کلمات کلیدی :

Brain–computer interface (BCI)،Electroencephalogram (EEG)،(Motor imagery (MI،)،Support vector machine (SVM)،Kernel-target alignment (KTA)،Embedded feature selection،l1 regularization،minibatch proximal gradient ascent

چکیده :

Brain-computer interface (BCI) based on motor imagery (MI) classification using scalp recorded multi-channel EEG signal play a major role in the control of artificial limbs and machines by people with severe disabilities. The most popular features are band power of EEG signals in frequency sub-bands in a relatively wide frequency band (e. g., 8-30 Hz) in different time windows of imagination period. The effectiveness of these features is highly dependent on the sub-bands and time windows adopted, since the optimal sub-bands and time windows is generally subject-specific. The nonlinear support vector machine (SVM) with Gaussian kernel is an excellent classifier for MI classification. In this paper, the SVM with an anisotropic Gaussian kernel with a special scaling factors for each feature is used, instead of one scale factor for all features. The scaling factors are tuned by maximizing the kernel-target alignment criterion with l1 regularization. Some of the scaling factors will be zero after maximization due to l1 regularization, that is equivalent to removing corresponding features from classification processes. Therefore, an embedded feature selection is also done to remove the destructive effect of irrelevant and redundant features. The average accuracy of the SVM classifier with anisotropic Gaussian kernel for four subjects in BCI IV-1 dataset is about 10% more than the SVM with isotropic Gaussian kernel.

لیست مقالات

لیست مقالات بایگانی شده

A Time-Based Analogue-to-Digital Converter for ECG Applications

Atiyeh Karimlou - Mohammad Yavari

Enhanced Current Commutation Drive Circuit for Hybrid DC Circuit Breaker

Alireza Jaafari - Sadegh Mohsenzade - Ali Asghar Razi-Kazemi

Improving Quarter-Wavelength Resonator Technique for Parasitic Cancellation of the ESD Protection Diode for High-Frequency Applications

Emadodin Zia Khodadadian - Mojtaba Joodaki

Family of Soft-Switched Single-Switch Switched-Resonator Converters with Low Component Count

Maryam Hajilou - Siamak Khalili - Hosein Farzanehfard

Net Load Forecasting of Household Prosumers Considering Deep Reinforcement Learning

Behzad Motallebi Azar - Rasool Kazemzadeh - Morteza Zare Oskouei - Behnam Mohammadi-Ivatloo

تاثیر روش‌های کاهش سناریو و عدم قطعیت‌های چندمتغیره بر عملکرد هاب انرژی

مهسا نعمتی فر - حسین شریف زاده

Power Consumption and I/Q-to-Phase Analysis in Direct Demodulation Approaches

Mir mahdi Safari - Jafar Pourrostam

Optimal Placement of Unified Power Flow Controller in Power System Considering Transient Stability and Voltage Stability Criteria

Esmail Zahmatkeshan - Mohsen Bandekhoda

Investigating Validity and Reliability of The Features Extracted by a 5R Vertical Robot for Arm Motion and Learning Assessment

Sarvenaz Bourbour - Fariba Bahrami Boodelalou - Ghorban Taghizadeh

Clustering of Fuzzy Data Based on Particle Swarm Optimization

Najme Ghanbari - Seyed-hamid Zahiri - Hadi Shahraki

بیشتر

ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 42.5.3