• Home
  • Search
  • Comparison of Domain Selection Methods for Multi-Source Manifold Feature Transfer Learning in Electroencephalogram Classification
  • Cite Icon1
  • https://doi.org/10.3390/app14062326Copy DOI Icon

Comparison of Domain Selection Methods for Multi-Source Manifold Feature Transfer Learning in Electroencephalogram Classification

Show More
  • Abstract
  • Highlights & Summary
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Transfer learning (TL) utilizes knowledge from the source domain (SD) to enhance the classification rate in the target domain (TD). It has been widely used to address the challenge of sessional and inter-subject variations in electroencephalogram (EEG)-based brain–computer interfaces (BCIs). However, utilizing knowledge from a combination of both related and non-related sources can significantly deteriorate the classification performance across individual target domains, resulting in a negative transfer (NT). Hence, NT becomes one of the most significant challenges for transfer learning algorithms. Notably, domain selection techniques have been developed to address the challenge of NT emerging from the transfer of knowledge from non-related sources. However, existing domain selection approaches iterate through domains and remove a single low-beneficial domain at a time, which can massively affect the classification performance in each iteration since SDs respond differently to other sources. In this paper, we compare domain selection techniques for a multi-source manifold feature transfer learning (MMFT) framework to address the challenge of NT and then evaluate the effect of beneficial and non-beneficial sources on TL performance. To evaluate the effect of low-beneficial and high beneficial sources on TL performance, some commonly used domain selection methods are compared, namely, domain transferability estimation (DTE), rank of domain (ROD), label similarity analysis, and enhanced multi-class MMFT (EMC-MMFT), using the same multi-class cross-session and cross-subject classification problems. The experimental results demonstrate the superiority of the EMC-MMFT algorithm in terms of minimizing the effect of NT. The highest classification accuracy (CA) of 100% is achieved when optimal combinations of high beneficial sources are selected for two-class SSMVEP problems, while the highest CAs of 98% and 87% are achieved for three- and four-class SSMVEP problems, respectively. The highest CA of 98% is achieved for two-class MI classification problems, while the highest CAs of 90% and 71.5% are obtained for three- and four-class MI problems, respectively.

Loading PDF

Similar Papers
  • Research Article
  • Citations14

Multi-source adversarial transfer learning for ultrasound image segmentation with limited similarity

  • Jul 26, 2023
  • Applied Soft Computing
  • Yifu Zhang +10
  • PDF
  • Research Article
  • Citations10

Multi-Class Transfer Learning and Domain Selection for Cross-Subject EEG Classification

  • Apr 21, 2023
  • Applied Sciences
  • Rito Clifford Maswanganyi +3
  • Research Article
  • Citations46

Multi-source manifold feature transfer learning with domain selection for brain-computer interfaces

  • Oct 05, 2022
  • Neurocomputing
  • Qingshan She +3
  • Research Article
  • Citations1

Multi-Source Transfer Learning With Spatial-Temporal Graph Neural Network for Short-Term Bicycle Traffic Prediction

  • Feb 01, 2026
  • IEEE Transactions on Intelligent Transportation Systems
  • Xiamei Wen +2
  • Research Article
  • Citations11

Multi-Source Selection Transfer Learning with Privacy-Preserving

  • May 07, 2022
  • Neural Processing Letters
  • Weifei Wu
  • Research Article
  • Citations21

Enhancing cross-subject EEG emotion recognition through multi-source manifold metric transfer learning

  • Apr 09, 2024
  • Computers in Biology and Medicine
  • Xinsheng Shi +5
  • PDF
  • Research Article
  • Citations43

Multi-Source Deep Transfer Neural Network Algorithm

  • Sep 16, 2019
  • Sensors (Basel, Switzerland)
  • Jingmei Li +3
  • Research Article
  • Citations84

A general multi-source ensemble transfer learning framework integrate of LSTM-DANN and similarity metric for building energy prediction

  • Sep 09, 2021
  • Energy and Buildings
  • Xi Fang +5
  • Research Article
  • Citations3

Single-Source and Multi-Source Cross-Subject Transfer Based on Domain Adaptation Algorithms for EEG Classification

  • Feb 27, 2025
  • Mathematics
  • Rito Clifford Maswanganyi +3
  • Research Article
  • Citations4

Multi-source sparse broad transfer learning for parkinson's disease diagnosis via speech.

  • Feb 04, 2025
  • Medical & biological engineering & computing
  • Yuchuan Liu +5
  • Research Article
  • Citations16

Multi-source variational mode transfer learning for enhanced PM2.5 concentration forecasting at data-limited monitoring stations

  • Sep 28, 2023
  • Expert Systems with Applications
  • Bozhi Yao +3
  • Research Article
  • Citations19

Multi-source fast transfer learning algorithm based on support vector machine.

  • Apr 06, 2021
  • Applied intelligence (Dordrecht, Netherlands)
  • Peng Gao +2
  • Conference Article
  • Citations16

Multi-source Transfer Learning with Ensemble for Financial Time Series Forecasting

  • Dec 01, 2020
  • Qi-Qiao He +2
  • Research Article
  • Citations2

Multisource Deep Transfer Learning Based on Balanced Distribution Adaptation

  • Apr 18, 2022
  • Computational Intelligence and Neuroscience
  • Peng Gao +3
  • PDF
  • Research Article
  • Citations15

A transfer learning model with multi-source domains for biomedical event trigger extraction

  • Jan 07, 2021
  • BMC Genomics
  • Yifei Chen
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.