A self-training semi-supervised SVM algorithm and its application in an EEG-based brain computer interface speller system

Li, Yuanqing; Guan, Cuntai; Li, Huiqi; Chin, Zheng Yang

doi:10.1016/j.patrec.2008.01.030

Cited by 200 publications

(129 citation statements)

References 15 publications

Supporting

Mentioning

123

Contrasting

Unclassified

Order By: Relevance

“…Li et al [25] propose a self-training semi-supervised support vector machine algorithm and a selection metric, which are designed for learning from a limited number of training data. Two examples show the validity of the algorithm and selection metric on a data set collected from a P300-based brain computer interface speller.…”

Section: Related Workmentioning

confidence: 99%

“…The main issue of semi-supervised learning is how to exploit information from the unlabeled data. A number of different algorithms for semi-supervised learning have been presented, such as the Expectation Maximization (EM) based algorithms [30,35], self-training [25,33,34,45], co-training [6,37], Transductive Support Vector Machine (TSVM) [23], SemiSupervised SVM (S3VM) [4], graph-based methods [2,48], and boosting based semi-supervised learning methods [27,38,40].…”

Section: Introductionmentioning

confidence: 99%

“…Bayesian methods, neural networks, logistic regression, marginbased classifiers, and algorithms that are normally seen as only outputting a classification model, like decision trees. Most of the current approaches to self-training utilize the first kind of learning algorithms as the base learner [25,33,34,45]. In this paper we focus on self-training with a decision tree learner as the base learner.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Semi-supervised self-training for decision tree classifiers

Tanha

Someren

Afsarmanesh

2015

Int. J. Mach. Learn. & Cyber.

207

View full text Add to dashboard Cite

We consider semi-supervised learning, learning task from both labeled and unlabeled instances and in particular, self-training with decision tree learners as base learners. We show that standard decision tree learning as the base learner cannot be effective in a self-training algorithm to semi-supervised learning. The main reason is that the basic decision tree learner does not produce reliable probability estimation to its predictions. Therefore, it cannot be a proper selection criterion in self-training. We consider the effect of several modifications to the basic decision tree learner that produce better probability estimation than using the distributions at the leaves of the tree. We show that these modifications do not produce better performance when used on the labeled data only, but they do benefit more from the unlabeled data in self-training. The modifications that we consider are Naive Bayes Tree, a combination of No-pruning and Laplace correction, grafting, and using a distance-based measure. We then extend this improvement to algorithms for ensembles of decision trees and we show that the ensemble learner gives an extra improvement over the adapted decision tree learners.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Semi-supervised self-training for decision tree classifiers

Tanha

Someren

Afsarmanesh

2015

Int. J. Mach. Learn. & Cyber.

207

View full text Add to dashboard Cite

show abstract

“…A great variety of classification algorithms was also used to design BCI systems. Linear Discriminant Analysis [18], Support Vector Machine (SVM) [19], and Hidden Markov Model [20] are some of those classifiers presented in the literature.…”

Section: Literature Reviewmentioning

confidence: 99%

EEG-based Brain-computer Interface for Automating Home Appliances

Alshbatat

Vial

Premaratne

et al. 2014

JCP

View full text Add to dashboard Cite

An EEG-based brain-computer system for automating home appliances is proposed in this study. Braincomputer interface (BCI) system provides direct pathway between human brain and external computing resources or external devices. The system translates thought into action without using muscles through a number of electrodes attached to the user's scalp. The BCI technology can be used by disabled people to improve their independence and maximize their capabilities at home. In this paper, a novel BCI system was developed to control home appliances from a dedicated Graphical User Interface (GUI). The system is structured with six units: EMOTIV EPOC headset, personal computer, Flyport module, quad band GSM/ GPRS communication module, LinkSprite JPEG Colour camera, and PIC-P40 board. EMOTIV EPOC headset detects and records neuronal electrical activities that reflect user's intent from different locations on the scalp. Those activities are then sent to the computer to extract specific signal features. Those features are then translated into commands to operate all appliances at home. The proposed system has been implemented, constructed, and tested. Experimental results demonstrates the feasibility of our proposed BCI system in controlling home appliances based on the user's physiological states. Abstract-An EEG-based brain-computer system for automating home appliances is proposed in this study. Braincomputer interface (BCI) system provides direct pathway between human brain and external computing resources or external devices. The system translates thought into action without using muscles through a number of electrodes attached to the user's scalp. The BCI technology

show abstract

“…There are also some approaches to reduce the calibration time using co-adaptive learning [9] or semi-supervised learning [10]. In these approaches, the BCI model is built first using very few signals from the new subject, and then it is adapted online using unsupervised or co-adaptive learning algorithms.…”

Section: Introductionmentioning

confidence: 99%

Subject-to-subject adaptation to reduce calibration time in motor imagery-based brain-computer interface

Arvaneh

Robertson

Ward

2014

2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

View full text Add to dashboard Cite

Abstract-In order to enhance the usability of a motor imagery-based brain-computer interface (BCI), it is highly desirable to reduce the calibration time. Due to inter-subject variability, typically a new subject has to undergo a 20-30 minutes calibration session to collect sufficient data for training a BCI model based on his/her brain patterns. This paper proposes a new subject-to-subject adaptation algorithm to reliably reduce the calibration time of a new subject to only 3-4 minutes. To reduce the calibration time, unlike several past studies, the proposed algorithm does not require a large pool of historic sessions. In the proposed algorithm, using only a few trials from the new subject, first, the new subject's data is adapted to each available historic session separately. This is done by a linear transformation minimizing the distribution difference between the two groups of EEG data. Thereafter, among the available historic sessions, the one matched the most to the new subject's adapted data is selected as the calibration session. Consequently, the previously trained model based on the selected historic session is entirely used for the classification of the new subject's data after adaptation. The proposed algorithm is evaluated on a publicly available dataset with 9 subjects. For each subject, the calibration session is selected only from the calibration sessions of the eight other subjects. The experimental results showed that our proposed algorithm not only reduced the calibration time by 85%, but also performed on average only 1.7% less accurate than the subject-dependent calibration results.

show abstract

A self-training semi-supervised SVM algorithm and its application in an EEG-based brain computer interface speller system

Cited by 200 publications

References 15 publications

Semi-supervised self-training for decision tree classifiers

Semi-supervised self-training for decision tree classifiers

EEG-based Brain-computer Interface for Automating Home Appliances

Subject-to-subject adaptation to reduce calibration time in motor imagery-based brain-computer interface

Contact Info

Product

Resources

About