How to measure the phase diffusion dynamics in the micromaser

We propose a realistic scheme for measuring the micromaser linewidth by monitoring the phase diffusion dynamics of the cavity field. Our strategy consists of exciting an initial coherent state with the same photon number distribution as the micromaser steady-state field, singling out a purely diffusive process in the system dynamics. After the injection of a counterfield, measurements of the population statistics of a probe atom allow us to derive the micromaser linewidth in all ranges of the relevant parameters, establishing experimentally the distinctive features of the micromaser spectrum due to the discreteness of the electromagnetic field.

Is a single item stress measure independently associated with subsequent severe injury: a prospective cohort study of 16,385 forest industry employees

Background: A previous review showed that high stress increases the risk of occupational injury by three- to five-fold. However, most of the prior studies have relied on short follow-ups. In this prospective cohort study we examined the effect of stress on recorded hospitalised injuries in an 8-year follow-up.
Methods: A total of 16,385 employees of a Finnish forest company responded to the questionnaire. Perceived stress was measured with a validated single-item measure, and analysed in relation recorded hospitalised injuries from 1986 to 2008. We used Cox proportional hazard regression models to examine the prospective associations between work stress, injuries and confounding factors.
Results: Highly stressed participants were approximately 40% more likely to be hospitalised due to injury over the follow-up period than participants with low stress. This association remained significant after adjustment for age, gender, marital status, occupational status, educational level, and physical work environment.
Conclusions: High stress is associated with an increased risk of severe injury.

Trying similarity, doing difference:The role of interviewer self-disclosure in interview talk with young people

Advocates of semi-structured interview techniques have often argued that rapport may be built, and power inequalities between interviewer and respondent counteracted, by strategic self-disclosure on the part of the interviewer. Strategies that use self-disclosure to construct similarity between interviewer and respondent rely on the presumption that the respondent will in fact interpret the interviewer's behaviour in this way. In this article we examine the role of interviewer self-disclosure using data drawn from three projects involving interviews with young people. We consider how an interviewer's attempts to ‘do similarity’ may be interpreted variously as displays of similarity or, ironically, as indicators of difference by the participant, and map the implications that this may have for subsequent interview dialogue. A particular object of concern relates to the ways in which self-disclosing acts may function in the negotiation of category entitlement within interview interactions.

A necessarily complex model to explain the biogeography of the amphibians and reptiles of Madagascar

Pattern and process are inextricably linked in biogeographic analyses, though we can observe pattern, we must infer process. Inferences of process are often based on ad hoc comparisons using a single spatial predictor. Here, we present an alternative approach that uses mixed-spatial models to measure the predictive potential of combinations of hypotheses. Biodiversity patterns are estimated from 8,362 occurrence records from 745 species of Malagasy amphibians and reptiles. By incorporating 18 spatially explicit predictions of 12 major biogeographic hypotheses, we show that mixed models greatly improve our ability to explain the observed biodiversity patterns. We conclude that patterns are influenced by a combination of diversification processes rather than by a single predominant mechanism. A ‘one-size-fits-all’ model does not exist. By developing a novel method for examining and synthesizing spatial parameters such as species richness, endemism and community similarity, we demonstrate the potential of these analyses for understanding the diversification history of Madagascar’s biota.

The validation of Fibit ZipTM physical activity monitor as a measure of free-living physical activity

BACKGROUND: The new generation of activity monitors allow users to upload their data to the internet and review progress. The aim of this study is to validate the Fitbit Zip as a measure of free-living physical activity.

FINDINGS: Participants wore a Fitbit Zip, ActiGraph GT3X accelerometer and a Yamax CW700 pedometer for seven days. Participants were asked their opinion on the utility of the Fitbit Zip. Validity was assessed by comparing the output using Spearman's rank correlation coefficients, Wilcoxon signed rank tests and Bland-Altman plots. 59.5% (25/47) of the cohort were female. There was a high correlation in steps/day between the Fitbit Zip and the two reference devices (r = 0.91, p < 0.001). No statistically significant difference between the Fitbit and Yamax steps/day was observed (Median (IQR) 7477 (3597) vs 6774 (3851); p = 0.11). The Fitbit measured significantly more steps/day than the Actigraph (7477 (3597) vs 6774 (3851); p < 0.001). Bland-Altman plots revealed no systematic differences between the devices.

CONCLUSIONS: Given the high level of correlation and no apparent systematic biases in the Bland Altman plots, the use of Fitbit Zip as a measure of physical activity. However the Fitbit Zip recorded a significantly higher number of steps per day than the Actigraph.

Similarity between Positronium-Atom and Electron-Atom Scattering

We employ the impulse approximation for a description of positronium-atom scattering. Our analysis and calculations of Ps-Kr and Ps-Ar collisions provide a theoretical explanation of the similarity between the cross sections for positronium scattering and electron scattering for a range of atomic and molecular targets observed by S. J. Brawley et al. [Science 330, 789 (2010)].

Connectivity and Physical Activity: Using Footpath Networks to Measure the Walkability of Built Environments

There is now a strong body of research that suggests that the form of the built environment can influence levels of physical activity, leading to an increasing interest in incorporating health objectives into spatial planning and regeneration policies and projects. There have been a number of strands to this research, one of which has sought to develop “objective” measurements of the built environment using Geographic Information Science (GIS) involving measures of connectivity and proximity to compare the relative “walkability” of different neighbourhoods. The development of the “walkability index” (e.g. Leslie et al 2007, Frank et al 2010) has become a popular indicator of spatial distribution of those features of the built environment that are considered to have the greatest positive influence on levels of physical activity. The success of this measure is built on its ability to succinctly capture built environment correlates of physical activity using routinely available spatial data, which includes using road centre lines as a basis of a proxy for connectivity.

This paper discusses two key aspects of the walkability index. First, it follows the suggestion of Chin et al (2008) that the use of a footpath network (where available), rather than road centre lines, may be far more effective in evaluating walkability. This may be particularly important for assessing changes in walkability arising from pedestrian-focused infrastructure projects, such as greenways. Second, the paper explores the implication of this for how connectivity can be measured. The paper takes six different measures of connectivity and first analyses the relationships between them and then tests their correlation with actual levels of physical activity of local residents in Belfast, Northern Ireland. The analysis finds that the best measurements appear to be intersection density and metric reach and uses this finding to discuss the implications of this for developing tools that may better support decision-making in spatial planning.

Extended Tree Augmented Naive Classifier

This work proposes an extended version of the well-known tree-augmented naive Bayes (TAN) classifier where the structure learning step is performed without requiring features to be connected to the class. Based on a modification of Edmonds’ algorithm, our structure learning procedure explores a superset of the structures that are considered by TAN, yet achieves global optimality of the learning score function in a very efficient way (quadratic in the number of features, the same complexity as learning TANs). A range of experiments show that we obtain models with better accuracy than TAN and comparable to the accuracy of the state-of-the-art classifier averaged one-dependence estimator.

A tree augmented classifier based on Extreme Imprecise Dirichlet Model

We present TANC, a TAN classifier (tree-augmented naive) based on imprecise probabilities. TANC models prior near-ignorance via the Extreme Imprecise Dirichlet Model (EDM). A first contribution of this paper is the experimental comparison between EDM and the global Imprecise Dirichlet Model using the naive credal classifier (NCC), with the aim of showing that EDM is a sensible approximation of the global IDM. TANC is able to deal with missing data in a conservative manner by considering all possible completions (without assuming them to be missing-at-random), but avoiding an exponential increase of the computational time. By experiments on real data sets, we show that TANC is more reliable than the Bayesian TAN and that it provides better performance compared to previous TANs based on imprecise probabilities. Yet, TANC is sometimes outperformed by NCC because the learned TAN structures are too complex; this calls for novel algorithms for learning the TAN structures, better suited for an imprecise probability classifier.

Bayesian network data imputation with application to survival tree analysis

Retrospective clinical datasets are often characterized by a relatively small sample size and many missing data. In this case, a common way for handling the missingness consists in discarding from the analysis patients with missing covariates, further reducing the sample size. Alternatively, if the mechanism that generated the missing allows, incomplete data can be imputed on the basis of the observed data, avoiding the reduction of the sample size and allowing methods to deal with complete data later on. Moreover, methodologies for data imputation might depend on the particular purpose and might achieve better results by considering specific characteristics of the domain. The problem of missing data treatment is studied in the context of survival tree analysis for the estimation of a prognostic patient stratification. Survival tree methods usually address this problem by using surrogate splits, that is, splitting rules that use other variables yielding similar results to the original ones. Instead, our methodology consists in modeling the dependencies among the clinical variables with a Bayesian network, which is then used to perform data imputation, thus allowing the survival tree to be applied on the completed dataset. The Bayesian network is directly learned from the incomplete data using a structural expectation–maximization (EM) procedure in which the maximization step is performed with an exact anytime method, so that the only source of approximation is due to the EM formulation itself. On both simulated and real data, our proposed methodology usually outperformed several existing methods for data imputation and the imputation so obtained improved the stratification estimated by the survival tree (especially with respect to using surrogate splits).

A Tree Augmented Classifier Based on Extreme Imprecise Dirichlet Model

In this paper we present TANC, i.e., a tree-augmented naive credal classifier based on imprecise probabilities; it models prior near-ignorance via the Extreme Imprecise Dirichlet Model (EDM) (Cano et al., 2007) and deals conservatively with missing data in the training set, without assuming them to be missing-at-random. The EDM is an approximation of the global Imprecise Dirichlet Model (IDM), which considerably simplifies the computation of upper and lower probabilities; yet, having been only recently introduced, the quality of the provided approximation needs still to be verified. As first contribution, we extensively compare the output of the naive credal classifier (one of the few cases in which the global IDM can be exactly implemented) when learned with the EDM and the global IDM; the output of the classifier appears to be identical in the vast majority of cases, thus supporting the adoption of the EDM in real classification problems. Then, by experiments we show that TANC is more reliable than the precise TAN (learned with uniform prior), and also that it provides better performance compared to a previous (Zaffalon, 2003) TAN model based on imprecise probabilities. TANC treats missing data by considering all possible completions of the training set, but avoiding an exponential increase of the computational times; eventually, we present some preliminary results with missing data.