Multi-Class Transfer Learning and Domain Selection for Cross-Subject EEG Classification
نویسندگان
چکیده
Transfer learning (TL) has been proven to be one of the most significant techniques for cross-subject classification in electroencephalogram (EEG)-based brain-computer interfaces (BCI). Hence, it is widely used address challenges cross-session and variability with more accurate intention prediction. In this case, TL utilizes knowledge (signal features) source domain(s) improve target domain. However, current existing transfer approaches on EEG-based BCI are mostly limited two-class problems, while multi-class problems only implemented a focus within-subject due complexity problems. paper, we first extended scenario, then investigated reason performance being poor classification. Secondly, challenge sessional subject-to-subject variations originating from both known unknown factors. It discovered that such have massive influence because negative (NT) across domains. Based discovery, propose approach based multi-source manifold feature (MMFT) framework an enhanced version minimize effects NT. The proposed extends MMFT cases. Then firstly searches domains high transferability selects best combination among (SD), utilize best-selected learning. Experimental results illustrate can employed three-class four-class also demonstrated could effectively effect significantly increase prediction rates individual (TD). highest accuracy (CA) 98% obtained by MMFT.
منابع مشابه
A subject transfer framework for EEG classification
This paper proposes a subject transfer framework for EEG classification. It aims to improve the classification performance when the training set of the target subject (namely user) is small owing to the need to reduce the calibration session. Our framework pursues improvement not only at the feature extraction stage, but also at the classification stage. At the feature extraction stage, we firs...
متن کاملCross-Subject EEG Feature Selection for Emotion Recognition Using Transfer Recursive Feature Elimination
Using machine-learning methodologies to analyze EEG signals becomes increasingly attractive for recognizing human emotions because of the objectivity of physiological data and the capability of the learning principles on modeling emotion classifiers from heterogeneous features. However, the conventional subject-specific classifiers may induce additional burdens to each subject for preparing mul...
متن کاملMulti Class Cross Domain Sentiment Classification Using Fuzzy Mapped Cluster
Nowadays, e-commerce is growing fast and customers make sure of quality based on product reviews. But the large no of product reviews makes it difficult to automatically classify them into different polarity classes (positive and negative). Most of the existing methods have come out for binary classification of customer reviews. The proposed work uses a fuzzy logic model for crossdomain sentime...
متن کاملMulti-class feature selection for texture classification
In this paper, a multi-class feature selection scheme based on recursive feature elimination (RFE) is proposed for texture classifications. The feature selection scheme is performed in the context of one-against-all least squares support vector machine classifiers (LSSVM). The margin difference between binary classifiers with and without an associated feature is used to characterize the discrim...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Applied sciences
سال: 2023
ISSN: ['2076-3417']
DOI: https://doi.org/10.3390/app13085205