FORCe: fully online and automated artifact removal for brain-computer interfacing

A fully automated and online artifact removal method for the electroencephalogram (EEG) is developed for use in brain-computer interfacing. The method (FORCe) is based upon a novel combination of wavelet decomposition, independent component analysis, and thresholding. FORCe is able to operate on a small channel set during online EEG acquisition and does not require additional signals (e.g. electrooculogram signals). Evaluation of FORCe is performed offline on EEG recorded from 13 BCI particpants with cerebral palsy (CP) and online with three healthy participants. The method outperforms the state-of the-art automated artifact removal methods Lagged auto-mutual information clustering (LAMIC) and Fully automated statistical thresholding (FASTER), and is able to remove a wide range of artifact types including blink, electromyogram (EMG), and electrooculogram (EOG) artifacts.

Cognitive behaviour therapy (CBT) for depression by computer vs. therapist: patient experiences and therapeutic processes

This case series compares patient experiences and therapeutic processes between two modalities of cognitive behaviour therapy (CBT) for depression: computerized CBT (cCBT) and therapist-delivered CBT (tCBT). In a mixed-methods repeated-measures case series, six participants were offered cCBT and tCBT in sequence, with the order of delivery randomized across participants. Questionnaires about patient experiences were administered after each session and a semi-structured interview was completed with each participant at the end of each therapy modality. Therapy expectations, patient experiences and session impact ratings in this study generally favoured tCBT. Participants typically experienced cCBT sessions as less meaningful, less positive and less helpful compared to tCBT sessions in terms of developing understanding, facilitating problem-solving and building a therapeutic relationship.

'You show me yours, I'll show you mine': the negotiation of shifts from textual to visual modes in computer-mediated interaction among gay men

This article explores the way users of an online gay chat room negotiate the exchange of photographs and the conduct of video conferencing sessions and how this negotiation changes the way participants manage their interactions and claim and impute social identities. Different modes of communication provide users with different resources for the control of information, affecting not just what users are able to reveal, but also what they are able to conceal. Thus, the shift from a purely textual mode for interacting to one involving visual images fundamentally changes the kinds of identities and relationships available to users. At the same time, the strategies users employ to negotiate these shifts of mode can alter the resources available in different modes. The kinds of social actions made possible through different modes, it is argued, are not just a matter of the modes themselves but also of how modes are introduced into the ongoing flow of interaction.

On the influence of imputation in classification: practical issues

The substitution of missing values, also called imputation, is an important data preparation task for many domains. Ideally, the substitution of missing values should not insert biases into the dataset. This aspect has been usually assessed by some measures of the prediction capability of imputation methods. Such measures assume the simulation of missing entries for some attributes whose values are actually known. These artificially missing values are imputed and then compared with the original values. Although this evaluation is useful, it does not allow the influence of imputed values in the ultimate modelling task (e.g. in classification) to be inferred. We argue that imputation cannot be properly evaluated apart from the modelling task. Thus, alternative approaches are needed. This article elaborates on the influence of imputed values in classification. In particular, a practical procedure for estimating the inserted bias is described. As an additional contribution, we have used such a procedure to empirically illustrate the performance of three imputation methods (majority, naive Bayes and Bayesian networks) in three datasets. Three classifiers (decision tree, naive Bayes and nearest neighbours) have been used as modelling tools in our experiments. The achieved results illustrate a variety of situations that can take place in the data preparation practice.

Evolutionary tuning of SVM parameter values in multiclass problems

Support vector machines (SVMs) were originally formulated for the solution of binary classification problems. In multiclass problems, a decomposition approach is often employed, in which the multiclass problem is divided into multiple binary subproblems, whose results are combined. Generally, the performance of SVM classifiers is affected by the selection of values for their parameters. This paper investigates the use of genetic algorithms (GAs) to tune the parameters of the binary SVMs in common multiclass decompositions. The developed GA may search for a set of parameter values common to all binary classifiers or for differentiated values for each binary classifier. (C) 2008 Elsevier B.V. All rights reserved.

HIERARCHICAL DECOMPOSITION OF MULTICLASS PROBLEMS

Several popular Machine Learning techniques are originally designed for the solution of two-class problems. However, several classification problems have more than two classes. One approach to deal with multiclass problems using binary classifiers is to decompose the multiclass problem into multiple binary sub-problems disposed in a binary tree. This approach requires a binary partition of the classes for each node of the tree, which defines the tree structure. This paper presents two algorithms to determine the tree structure taking into account information collected from the used dataset. This approach allows the tree structure to be determined automatically for any multiclass dataset.

Poly-bagging predictors for classification modelling for credit scoring

Credit scoring modelling comprises one of the leading formal tools for supporting the granting of credit. Its core objective consists of the generation of a score by means of which potential clients can be listed in the order of the probability of default. A critical factor is whether a credit scoring model is accurate enough in order to provide correct classification of the client as a good or bad payer. In this context the concept of bootstraping aggregating (bagging) arises. The basic idea is to generate multiple classifiers by obtaining the predicted values from the fitted models to several replicated datasets and then combining them into a single predictive classification in order to improve the classification accuracy. In this paper we propose a new bagging-type variant procedure, which we call poly-bagging, consisting of combining predictors over a succession of resamplings. The study is derived by credit scoring modelling. The proposed poly-bagging procedure was applied to some different artificial datasets and to a real granting of credit dataset up to three successions of resamplings. We observed better classification accuracy for the two-bagged and the three-bagged models for all considered setups. These results lead to a strong indication that the poly-bagging approach may promote improvement on the modelling performance measures, while keeping a flexible and straightforward bagging-type structure easy to implement. (C) 2011 Elsevier Ltd. All rights reserved.

A Road Following Approach Using Artificial Neural Networks Combinations

Navigation is a broad topic that has been receiving considerable attention from the mobile robotic community over the years. In order to execute autonomous driving in outdoor urban environments it is necessary to identify parts of the terrain that can be traversed and parts that should be avoided. This paper describes an analyses of terrain identification based on different visual information using a MLP artificial neural network and combining responses of many classifiers. Experimental tests using a vehicle and a video camera have been conducted in real scenarios to evaluate the proposed approach.

Computer simulations of copper and gold nanowires and single-wall nanowires

Copper and gold nanowires under tension evolve to form linear atomic chains (LACs), and the study and understanding of this evolution is an important subject for the development of nanocontacts. Here we study the differences and similarities between copper and gold nanowires (NWs) under stress along the [111] crystallographic direction until their rupture using tight-binding molecular dynamics. In both metals, the first significant rearrangement occurs due to one inside atom that goes to the NW` surface. In an attempt to better understand this effect, for both metals we also consider hollow NW`s where the inside atoms were excluded after the initial relaxation to create single-wall NW`s (SWNWs). The dynamical evolution of these SWNWs provides insight on the formation of the constriction that evolves to form LACs. Studying the calculated forces supported by the NW`s we show that SWNWs can sustain larger forces before the first major rearrangement in the copper and gold when compared to the original NW`s.

A novel MAP-MRF approach for multispectral image contextual classification using combination of suboptimal iterative algorithms

In this paper we present a novel approach for multispectral image contextual classification by combining iterative combinatorial optimization algorithms. The pixel-wise decision rule is defined using a Bayesian approach to combine two MRF models: a Gaussian Markov Random Field (GMRF) for the observations (likelihood) and a Potts model for the a priori knowledge, to regularize the solution in the presence of noisy data. Hence, the classification problem is stated according to a Maximum a Posteriori (MAP) framework. In order to approximate the MAP solution we apply several combinatorial optimization methods using multiple simultaneous initializations, making the solution less sensitive to the initial conditions and reducing both computational cost and time in comparison to Simulated Annealing, often unfeasible in many real image processing applications. Markov Random Field model parameters are estimated by Maximum Pseudo-Likelihood (MPL) approach, avoiding manual adjustments in the choice of the regularization parameters. Asymptotic evaluations assess the accuracy of the proposed parameter estimation procedure. To test and evaluate the proposed classification method, we adopt metrics for quantitative performance assessment (Cohen`s Kappa coefficient), allowing a robust and accurate statistical analysis. The obtained results clearly show that combining sub-optimal contextual algorithms significantly improves the classification performance, indicating the effectiveness of the proposed methodology. (C) 2010 Elsevier B.V. All rights reserved.

Computer-aided Structure Elucidation of Neolignans

This paper describes a new module of the expert system SISTEMAT used for the prediction of the skeletons of neolignans by (13)C NMR, (1)H NMR and botanical data obtained from the literature. SISTEMAT is composed of MACRONO, SISCONST, C13MACH, H1MACH and SISOCBOT programs, each analyzing data of the neolignan in question to predict the carbon skeleton of the compound. From these results, the global probability is computed and the most probable skeleton predicted. SISTEMAT predicted the skeletons of 75% of the 20 neolignans tested, in a rapid and simple procedure demonstrating its advantage for the structural elucidation of new compounds.

Transport coefficients, Raman spectroscopy, and computer simulation of lithium salt solutions in an ionic liquid

Lithium salt solutions of Li(CF3SO2)(2)N, LiTFSI, in a room-temperature ionic liquid (RTIL), 1-butyl-2,3-dimethyl-imidazolium cation, BMMI, and the (CF3SO2)(2)N-, bis(trifluoromethanesulfonyl)imide anion, [BMMI][TFSI], were prepared in different concentrations. Thermal properties, density, viscosity, ionic conductivity, and self-diffusion coefficients were determined at different temperatures for pure [BMMI][TFSI] and the lithium solutions. Raman spectroscopy measurements and computer simulations were also carried out in order to understand the microscopic origin of the observed changes in transport coefficients. Slopes of Walden plots for conductivity and fluidity, and the ratio between the actual conductivity and the Nernst-Einstein estimate for conductivity, decrease with increasing LiTFSI content. All of these studies indicated the formation of aggregates of different chemical nature, as it is corroborated by the Raman spectra. In addition, molecular dynamics (MD) simulations showed that the coordination of Li+ by oxygen atoms of TFSI anions changes with Li+ concentration producing a remarkable change of the RTIL structure with a concomitant reduction of diffusion coefficients of all species in the solutions.

Computer-Assisted Approach to Structural Elucidation of Lignans

This paper reports an expert system (SISTEMAT) developed for structural determination of diverse chemical classes of natural products, including lignans, based mainly on 13C NMR and 1H NMR data of these compounds. The system is composed of five programs that analyze specific data of a lignan and shows a skeleton probability for the compound. At the end of analyses, the results are grouped, the global probability is computed, and the most probable skeleton is exhibited to the user. SISTEMAT was able to properly predict the skeletons of 80% of the 30 lignans tested, demonstrating its advantage during the structural elucidation course in a short period of time.