Kernel PCA for Feature Extraction and De-Noising in Nonlinear Regression

In this paper, we propose the application of the Kernel Principal Component Analysis (PCA) technique for feature selection in a high-dimensional feature space, where input variables are mapped by a Gaussian kernel. The extracted features are employed in the regression problems of chaotic Mackey–Glass time-series prediction in a noisy environment and estimating human signal detection performance from brain event-related potentials elicited by task relevant signals. We compared results obtained using either Kernel PCA or linear PCA as data preprocessing steps. On the human signal detection task, we report the superiority of Kernel PCA feature extraction over linear PCA. Similar to linear PCA, we demonstrate de-noising of the original data by the appropriate selection of various nonlinear principal components. The theoretical relation and experimental comparison of Kernel Principal Components Regression, Kernel Ridge Regression and ε-insensitive Support Vector Regression is also provided.     

A web-based e-learning system for increasing study efficiency by stimulating learner’s motivation

Due to the opportunities provided by the Internet, more and more people are taking advantage of distance learning courses and during the last few years enormous research efforts have been dedicated to the development of distance learning systems. So far, many e-learning systems are proposed and used practically. However, in these systems the e-learning completion rate is about 30%. One of the reasons is the low study desire when the learner studies the learning materials. In this research, we propose an interactive Web-based e-learning system. The purpose of our system is to increase the e-learning completion rate by stimulating learner’s motivation. The proposed system has three subsystems: the learning subsystem, learner support subsystem, and teacher support subsystem. The learning subsystem improves the learner’s study desire. The learner support subsystem supports the learner during the study, and the teacher support subsystem supports the teacher to get the learner’s study state. To evaluate the proposed system, we developed several experiments and surveys. By using new features such as: display of learner’s study history, change of interface color, encourage function, ranking function, self-determination of the study materials, and grouping of learners, the proposed system can increase the learning efficiency.

Local search methods for efficient router nodes placement in wireless mesh networks

Wireless Mesh Networks (WMNs) are an important networking infrastructure for providing cost-efficient broadband wireless connectivity to a group of users. WMNs are increasingly being used in urban, metropolitan and municipal area networks for deployment of medical, transport, surveillance systems, etc. The performance and operability of WMNs largely depends on placement of mesh routers nodes in the geographical area to achieve network connectivity and stability. The objective is to find an optimal and robust topology of the mesh router network to support intelligent telecommunication services to clients such as adaptive and flexible wireless Internet access, mobile data, voice, video in addition to supporting other preferred client services. In this work, we propose and evaluate local search methods for intelligent placement of mesh routers in WMNs with a two fold objective: maximizing the size of the giant component in the network and user coverage. Given a grid area where to distribute a given number of mesh router nodes, which can have different radio coverage, and a number of fixed clients a priori distributed in the given area, local search methods explore different local movements and incrementally improve the quality of the router nodes placement in terms of network connectivity and user coverage. We have experimentally evaluated the proposed local search methods through a benchmark of generated instances of varying sizes. Moreover, different distributions of mesh clients (Uniform, Normal, Exponential and Weibull) are used. The experimental evaluation showed the good performance of local search methods for optimizing network connectivity and user coverage in WMNs.     

A medical hand tool physical interaction evaluation approach for prototype testing using patient care simulators

The purpose of this study was to develop and test a physical ergonomics assessment approach for medical device handles. The method assesses wrist posture and force of exertion simultaneously by task element. Electrogoniometers and EMG sensors were connected through a data acquisition module for synchronization with video recordings of trials. Task analysis of video recordings were performed offline with Observer XT software. Average posture and force data across several repetitions of individual task elements were calculated and presented in a format suitable for informing product designers of specific issues during a test trial. A handle comfort questionnaire is proposed to survey subjective responses. The evaluation approach was applied to an endoscope needle in sampling a biopsy from the stomach wall using a gastrointestinal track simulator with ten physician surrogates. The results showed that for all task elements the wrist was in extension (33(°)-45(°)). Peak muscle forces ranged from 28% to 68% MVC across the three muscles studied. Muscle peak forces were above ACGIH HAL maximum threshold limits for four of the seven task elements, and above the action limit for all seven task elements for two muscles. The handle comfort questionnaire data also supported the high muscle force findings, and also on force distribution on the handle due to contact stresses. This combined approach could be used to collect and report detailed early stage ergonomics data from user trials on patient care simulators. The approach is proposed for use by medical device designers at the design stage of new products using prototypes, but it could also be used on existing products with real patients.     

Brain-computer interfaces for 1-D and 2-D cursor control: designs using volitional control of the EEG spectrum or steady-state visual evoked potentials.

We have developed and tested two electroencephalogram (EEG)-based brain-computer interfaces (BCI) for users to control a cursor on a computer display. Our system uses an adaptive algorithm, based on kernel partial least squares classification (KPLS), to associate patterns in multichannel EEG frequency spectra with cursor controls. Our first BCI, Target Practice, is a system for one-dimensional device control, in which participants use biofeedback to learn voluntary control of their EEG spectra. Target Practice uses a KPLS classifier to map power spectra of 62-electrode EEG signals to rightward or leftward position of a moving cursor on a computer display. Three subjects learned to control motion of a cursor on a video display in multiple blocks of 60 trials over periods of up to six weeks. The best subject's average skill in correct selection of the cursor direction grew from 58% to 88% after 13 training sessions. Target Practice also implements online control of two artifact sources: 1) removal of ocular artifact by linear subtraction of wavelet-smoothed vertical and horizontal electrooculograms (EOG) signals, 2) control of muscle artifact by inhibition of BCI training during periods of relatively high power in the 40-64 Hz band. The second BCI, Think Pointer, is a system for two-dimensional cursor control. Steady-state visual evoked potentials (SSVEP) are triggered by four flickering checkerboard stimuli located in narrow strips at each edge of the display. The user attends to one of the four beacons to initiate motion in the desired direction. The SSVEP signals are recorded from 12 electrodes located over the occipital region. A KPLS classifier is individually calibrated to map multichannel frequency bands of the SSVEP signals to right-left or up-down motion of a cursor on a computer display. The display stops moving when the user attends to a central fixation point. As for Target Practice, Think Pointer also implements wavelet-based online removal of ocular artifact; however, in Think Pointer muscle artifact is controlled via adaptive normalization of the SSVEP. Training of the classifier requires about 3 min. We have tested our system in real-time operation in three human subjects. Across subjects and sessions, control accuracy ranged from 80% to 100% correct with lags of 1-5 s for movement initiation and turning. We have also developed a realistic demonstration of our system for control of a moving map display (http://ti.arc.nasa.gov/).     

The Occurrence of Male-to-Female Intimate Partner Violence on Days of Men's Drinking: The Moderating Effects of Antisocial Personality Disorder.

In this study, the moderating effects of antisocial personality disorder (ASPD) on the day-to-day relationship between male partner alcohol consumption and male-to-female intimate partner violence (IPV) for men entering a domestic violence treatment program (n = 170) or an alcoholism treatment program (n = 169) were examined. For both samples, alcohol consumption was associated with an increased likelihood of nonsevere IPV among men without a diagnosis of ASPD but not among men with ASPD (who tended to engage in nonsevere IPV whether they did or did not drink). Drinking was more strongly associated with a likelihood of severe IPV among men with ASPD compared with those without ASPD who also drank. These results provide partial support for a multiple threshold model of intoxication and aggression. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A theory of pad conditioning for chemical-mechanical polishing

Statistical models are presented to describe the evolution of the surface roughness of polishing pads during the pad-conditioning process in chemical-mechanical polishing. The models describe the evolution of the surface-height probability-density function of solid pads during fixed height or fixed cut-rate conditioning. An integral equation is derived for the effect of conditioning on a foamed pad in terms of a model for a solid pad. The models that combine wear and conditioning are then discussed for both solid and foamed pads. Models include the dependence of the surface roughness on the shape and density of the cutting tips used in the conditioner and on other operating parameters. Good agreement is found between the model, Monte Carlo simulations and with experimental data.     

Memory and Emotions for the September 11, 2001, Terrorist Attacks in Patients With Alzheimer's Disease, Patients With Mild Cognitive Impairment, and Healthy Older Adults.

National traumatic events can produce extremely vivid memories. Using a questionnaire administered during telephone interviews, the authors investigated emotional responses to, and memory for. the September 11, 2001, terrorist attacks in patients with Alzheimer's disease (AD), patients with mild cognitive impairment (MCI), and healthy older adults in the initial weeks following the event and again 3-4 months later. There were several notable findings. First, patients with AD showed less memory than patients with MCI and older adults. Second, patients with AD, but not patients with MCI or older adults, appeared to retain more memory for personal versus factual information. Third, patients with AD and older adults did not differ in the intensity of their reported emotional responses to the attacks, whereas patients with MCI reported relatively less intense emotional responses. Last, distortions of memory for personal information were frequent for all participants but were more common in patients with AD. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A Transactional Model of Parent-Infant Interactions in Alcoholic Families.

This study examined the transactional nature of parent-infant interactions over time among alcoholic and nonalcoholic families. The sample consisted of 222 families assessed at 12, 18, and 24 months of child age. Results indicated that infant behavior did not influence parental behavior across time, but parental behavior was longitudinally predictive of infant behavior during play interactions. Higher paternal alcohol consumption at 12 months was longitudinally predictive of negative parental behavior at 24 months. Other significant risk factors included marital conflict, fathers' depression, and fathers' education. Results highlight the nested nature of risk in alcoholic families and the direction of influence from parent to child during interactions and suggest that 1 pathway to risk among these children is through negative parent-infant interactions. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Determining the absolute efficiency of a delay line microchannel-plate detector using molecular dissociation

We present a method to measure the absolute detection efficiency of a delay-line microchannel-plate detector using the breakup of diatomic molecular ions. This method provides the absolute total detection efficiency, as well as the individual efficiency for each signal of the detector. The method is based on the fact that molecular breakup always yields two hits on the detector, but due to finite detection efficiency some of these events are recorded as single particles while others are detected in pairs. We demonstrate the method by evaluating the detection efficiency for both timing and position signals of a delay-line detector using laser-induced dissociation of molecular ions. In addition, the detection efficiency as a function of position has been determined by dividing the detector into sectors.