Shang-Wen Chuang

Cluster-based support vector machines in text-independent speaker identification

Based on statistical learning theory, support vector machines (SVM) is a powerful tool for various classification problems, such as pattern recognition and speaker identification etc. However, training SVM consumes large memory and long computing time. This work proposes a cluster-based learning methodology to reduce training time and the memory size for SVM. By using k-means based clustering technique, training data at boundary of each cluster were selected for SVM learning. We also applied this technique to text-independent speaker identification problems. Without deteriorating recognition performance, the training data and time can be reduced up to 75% and 87.5% respectively.

EEG Effects of Motion Sickness Induced in a Dynamic Virtual Reality Environment

The Electroencephalogram (EEG) dynamics which relate to motion sickness are studied in this paper. Instead of providing visual or motion stimuli to the subjects to induce motion sickness, we employed a dynamic virtual-reality (VR) environment in our research. The environment consisted of a 3D surrounding VR scene and a motion platform providing a realistic situation. This environment provided the advantages of safety, low cost, and the realistic stimuli to induce motion sickness. The Motion Sickness Questionnaire (MSQ) was used to assess the sickness level, and the EEG effects on the subjects with high sickness levels were investigated using the independent component analysis (ICA). The fake-epoch extraction was then applied to the nausea-related independent components. Finally we employed the Event-Related Spectral Perturbation (ERSP) technology on the fake-epochs in order to determine the EEG dynamics during motion sickness. The experimental results show that most subjects experienced an 8-10 Hz power increase to their motion sickness-related phenomena in the parietal and motor areas. Moreover, some subjects experienced an EEG power increase of 18-20 Hz in their synchronized responses recorded in the same areas. The motion sickness-related effects and regions can be successfully obtained from our experimental results.

Implementation of a motion sickness evaluation system based on EEG spectrum analysis

Motion sickness is a normal response to real, perceived, or even anticipated movement. People tend to get motion sickness on a moving boat, train, airplane, car, or amusement park rides. Motion sickness occurs when the body, the inner ear, and the eyes send conflicting signals to the brain. Sensory conflict theory that came about in the 1970s has become the most widely accepted theorem of motion-sickness among scientists [1]. The theory proposed that the conflict between the incoming sensory inputs could induce motion-sickness. However, some new research studies have appeared to tackle the issue of the vestibular function in central nervous system (CNS). In the previous human subject studies, researchers attempt to confirm the brain areas involved in the conflict in multi-modal sensory systems by means of clinical or anatomical methods. Our past studies had investigated the EEG activities correlated with motion sickness in a virtual-reality based driving simulator. We found that the parietal, motor, occipital brain regions exhibited significant EEG power changes in response to vestibular and visual stimuli. Based on these experimental results, we attempt to implement an EEG-based evaluation system to estimate subjects motion sickness level upon the major EEG power spectra from these motion sickness related brain area in this study. The evaluation system can be applied to early detect the subjects motion sickness level and prevent the uncomfortable syndromes occurred in advance in our daily life.

EEG Alpha and Gamma Modulators Mediate Motion Sickness-Related Spectral Responses

Motion sickness (MS) is a common experience of travelers. To provide insights into brain dynamics associated with MS, this study recruited 19 subjects to participate in an electroencephalogram (EEG) experiment in a virtual-reality driving environment. When riding on consecutive winding roads, subjects experienced postural instability and sensory conflict between visual and vestibular stimuli. Meanwhile, subjects rated their level of MS on a six-point scale. Independent component analysis (ICA) was used to separate the filtered EEG signals into maximally temporally independent components (ICs). Then, reduced logarithmic spectra of ICs of interest, using principal component analysis, were decomposed by ICA again to find spectrally fixed and temporally independent modulators (IMs). Results demonstrated that a higher degree of MS accompanied increased activation of alpha (r = 0.421) and gamma (r =0.478) IMs across remote-independent brain processes, covering motor, parietal and occipital areas. This co-modulatory spectral change in alpha and gamma bands revealed the neurophysiological demand to regulate conflicts among multi-modal sensory systems during MS.

Genetic feature selection in EEG-based motion sickness estimation

Motion sickness is a common symptom that occurs when the brain receives conflicting information about the sensation of movement. Many motion sickness biomarkers have been identified, and electroencephalogram (EEG)-based motion sickness level estimation was found feasible in our previous study. This study employs genetic feature selection to find a subset of EEG features that can further improve estimation performance over the correlation-based method reported in the previous studies. The features selected by genetic feature selection were very different from those obtained by correlation analysis. Results of this study demonstrate that genetic feature selection is a very effective method to optimize the estimation of motion-sickness level. This demonstration could lead to a practical system for noninvasive monitoring of the motion sickness of individuals in real-world environments.

Forecasting electricity market prices: a neural network based approach

This work presents a neural network approach to forecast the Phelix Base (PB) electricity market prices for European Energy Exchange (EEX). Up to now there has been little scientific work on forecasting the price development on the electricity markets. In this study, the Phelix Base moving average (PBMA), the moving difference (PBMD), and multilayer feedforward neural networks (MLNN) are used to predict various period for 7, 14, 21, 28, 63, 91, 182, and 273 days ahead of electric prices. The experimental results of forecasting by MLNNs and linear methods (autoregressive error model) are compared and discussed. The MLNNs outperform from 11.4% to 64.6% superior to the traditional linear regression method. It seems that the proposed MLNN can be very useful in predicting the electricity market prices of EEX.

Independent modulators mediate spectra of multiple brain processes in a VR-based driving experiment

This study explores the use of Independent Component Analysis (ICA) applied to normalized logarithmic spectral changes in the activities of brain processes separated by spatial filters learned from electroencephalogram (EEG) data using a temporal ICA. EEG data were collected during 1-2 hour virtual-reality based driving experiments, in which subjects were instructed to maintain their cruising position and compensate for randomly induced drifts using the steering wheel. ICA was first applied to 30-channel EEG data to separate the recorded signals into a sum of maximally temporally independent components (ICs) for each of 15 subjects. Logarithmic spectra of IC activities were then submitted to PCA-ICA to find spectrally fixed and temporally independent modulator (IM) processes. The second ICA detected and modeled independent co-modulatory systems that multiplicatively affect the activities of spatially distinct IC processes. Across subjects, we found two consistent temporally independent modulators: theta-beta and alpha modulators that mediate spectral activations of the distinct cortical areas when the participants experience waves of alternating alertness and drowsiness during long hour simulated driving. Furthermore, the time courses of the theta-beta modulator were highly correlated with concurrent changes in subject driving error (a behavioral index of drowsiness).

Motion-Sickness Related Brain Areas and EEG Power Activates

This study investigates electroencephalographic (EEG) correlates of motion sickness in a virtual-reality based driving simulator. The driving simulator comprised an actual automobile mounted on a Stewart motion platform with six degrees of freedom, providing both visual and vestibular stimulations to induce motion-sickness in a manner that is close to that in daily life. EEG data were acquired at a sampling rate of 500 Hz using a 32-channel EEG system. The acquired EEG signals were analyzed using independent component analysis (ICA) and time-frequency analysis to assess EEG correlates of motion sickness. Subjects degree of motion-sickness was simultaneously and continuously reported using an onsite joystick, providing non-stop psychophysical references to the recorded EEG changes. Five Motion-sickness related brain processes with equivalent dipoles located in the left motor, the parietal, the right motor, the occipital and the occipital midline areas were consistently identified across all subjects. These components exhibited distinct spectral suppressions or augmentation in motion sickness. The results of this study could lead to a practical human-machine interface for noninvasive monitoring of motion sickness of drivers or passengers in real-world environments.