Root Cause Analysis by a Combined Sparse Classification and Monte Carlo Approach

Classification methods with embedded feature selection capability are very appealing for the analysis of complex processes since they allow the analysis of root causes even when the number of input variables is high. In this work, we investigate the performance of three techniques for classification within a Monte Carlo strategy with the aim of root cause analysis. We consider the naive bayes classifier and the logistic regression model with two different implementations for controlling model complexity, namely, a LASSO-like implementation with a L1 norm regularization and a fully Bayesian implementation of the logistic model, the so called relevance vector machine. Several challenges can arise when estimating such models mainly linked to the characteristics of the data: a large number of input variables, high correlation among subsets of variables, the situation where the number of variables is higher than the number of available data points and the case of unbalanced datasets. Using an ecological and a semiconductor manufacturing dataset, we show advantages and drawbacks of each method, highlighting the superior performance in term of classification accuracy for the relevance vector machine with respect to the other classifiers. Moreover, we show how the combination of the proposed techniques and the Monte Carlo approach can be used to get more robust insights into the problem under analysis when faced with challenging modelling conditions.

Robust visual tracking using structurally random projection and weighted least squares

Sparse representation based visual tracking approaches have attracted increasing interests in the community in recent years. The main idea is to linearly represent each target candidate using a set of target and trivial templates while imposing a sparsity constraint onto the representation coefficients. After we obtain the coefficients using L1-norm minimization methods, the candidate with the lowest error, when it is reconstructed using only the target templates and the associated coefficients, is considered as the tracking result. In spite of promising system performance widely reported, it is unclear if the performance of these trackers can be maximised. In addition, computational complexity caused by the dimensionality of the feature space limits these algorithms in real-time applications. In this paper, we propose a real-time visual tracking method based on structurally random projection and weighted least squares techniques. In particular, to enhance the discriminative capability of the tracker, we introduce background templates to the linear representation framework. To handle appearance variations over time, we relax the sparsity constraint using a weighed least squares (WLS) method to obtain the representation coefficients. To further reduce the computational complexity, structurally random projection is used to reduce the dimensionality of the feature space while preserving the pairwise distances between the data points in the feature space. Experimental results show that the proposed approach outperforms several state-of-the-art tracking methods.

Seasonal variation in month of diagnosis in children with type 1 diabetes registered in 23 European centers during 1989-2008: little short-term influence of sunshine hours or average temperature

BACKGROUND: The month of diagnosis in childhood type 1 diabetes shows seasonal variation.

OBJECTIVE: We describe the pattern and investigate if year-to-year irregularities are associated with meteorological factors using data from 50 000 children diagnosed under the age of 15 yr in 23 population-based European registries during 1989-2008.

METHODS: Tests for seasonal variation in monthly counts aggregated over the 20 yr period were performed. Time series regression was used to investigate if sunshine hour and average temperature data were predictive of the 240 monthly diagnosis counts after taking account of seasonality and long term trends.

RESULTS: Significant sinusoidal pattern was evident in all but two small centers with peaks in November to February and relative amplitudes ranging from ±11 to ±38% (median ±17%). However, most centers showed significant departures from a sinusoidal pattern. Pooling results over centers, there was significant seasonal variation in each age-group at diagnosis, with least seasonal variation in those under 5 yr. Boys showed greater seasonal variation than girls, particularly those aged 10-14 yr. There were no differences in seasonal pattern between four 5-yr sub-periods. Departures from the sinusoidal trend in monthly diagnoses in the period were significantly associated with deviations from the norm in average temperature (0.8% reduction in diagnoses per 1 °C excess) but not with sunshine hours.

CONCLUSIONS: Seasonality was consistently apparent throughout the period in all age-groups and both sexes, but girls and the under 5 s showed less marked variation. Neither sunshine hour nor average temperature data contributed in any substantial way to explaining departures from the sinusoidal pattern.

Pooled analysis of tiotropium Respimat® pharmacokinetics in cystic fibrosis

Tiotropium is the first bronchodilator to be studied systematically in cystic fibrosis (CF). We investigated whether any intrinsic or extrinsic factors affected pharmacokinetic (PK) parameters of inhaled tiotropium delivered by Respimat® in adults and children with CF. Tiotropium PK in patients with CF was compared with that of healthy volunteers and patients with chronic obstructive pulmonary disease (COPD). This pooled analysis summarizes the PK parameters of inhaled tiotropium Respimat® across 9 early- and late-phase trials involving 27 healthy volunteers (1 trial), 409 patients with CF (3 trials), and 281 patients with COPD (5 trials). Patients with CF aged 5 to 11, 12 to 17, and ≥18 years had similar tiotropium plasma concentrations (geometric mean C_{0.083,ss,norm}: 2.22pg/mL/μg; not determined for patients aged <5 years). The fraction excreted unchanged in the urine was 3.4-fold lower for patients aged 0.4 to <5 years than for those aged 5 to 11 years (fe_0-4,ss: 1.19% vs 4.09%). Tiotropium PK parameters were similar between CF patients and COPD patients.

Norms of basic elementary operators on algebras of regular operators

We show that if E is an atomic Banach lattice with an ordercontinuous norm, A, B ∈ L^r(E) and M_A,_B is the operator on L^r(E) defined by M_A,_B(T) = AT B then ||M_A,_B||r = ||A||_r||B||_r but that there is no real α > 0 such that ||MA,B || ≥ α ||A||r||B ||r.

Two dimensional unital Riesz algebras, their representations and norms

We describe all two dimensional unital Riesz algebras and study representations of them in Riesz algebras of regular operators. Although our results are not complete, we do demonstrate that very varied behaviour can occur even though all these algebras can be given a Banach lattice algebra norm.

A novel forward gene selection algorithm for microarray data

This paper investigates the gene selection problem for microarray data with small samples and variant correlation. Most existing algorithms usually require expensive computational effort, especially under thousands of gene conditions. The main objective of this paper is to effectively select the most informative genes from microarray data, while making the computational expenses affordable. This is achieved by proposing a novel forward gene selection algorithm (FGSA). To overcome the small samples' problem, the augmented data technique is firstly employed to produce an augmented data set. Taking inspiration from other gene selection methods, the L2-norm penalty is then introduced into the recently proposed fast regression algorithm to achieve the group selection ability. Finally, by defining a proper regression context, the proposed method can be fast implemented in the software, which significantly reduces computational burden. Both computational complexity analysis and simulation results confirm the effectiveness of the proposed algorithm in comparison with other approaches

In Deutschland mehr Vorbehalte gegenüber älteren Arbeitnehmern: ein deutsch-britischer Vergleich

Infolge steigender Lebenserwartung sowie sinkender Geburtenraten wird das Durchschnittsalter der Bevölkerung in den meisten Industrienationen in den kommenden 50 Jahren stark ansteigen. Zur gleichen Zeit scheidet nach wie vor die Mehrzahl älterer Arbeitnehmer vor Erreichen der gesetzlichen Altersgrenze aus dem Arbeitsmarkt aus. Dies hat sowohl negative Auswirkungen auf die Finanzierbarkeit sozialer Sicherungssysteme als auch langfristig auf die Verfügbarkeit qualifizierter Arbeitnehmer am Arbeitsmarkt. Nationale Regierungen verfolgen daher gegenwärtig verschiedene Maßnahmen mit dem Ziel, die Arbeitsmarktbeteiligung älterer Arbeitnehmer zu erhöhen (OECD 2006). Diese Maßnahmen werden jedoch zum Teil durch die Strategien des betrieblichen Personalmanagements konterkariert, die oft noch jugendzentristisch ausgeprägt sind und ältere Mitarbeiter bei Einstellung, Beförderung, Weiterbildung sowie bei Arbeitsplatz- und Arbeitszeitgestaltung direkt oder indirekt benachteiligen (Bellmann et al. 2003).

The Disorganized Family: Institutions, Practices and Normativity

This paper considers the value of a normative account of the relationship between agents and institutions for contemporary efforts to explain ever more complex and disorganized forms of social life. The character of social institutions, as they relate to practices, agents and norms, is explored through an engagement with the common claim that family life has been deinstitutionalized. The paper argues that a normative rather than empirical definition of institutions avoids a false distinction between institutions and practices. Drawing on ideas of social freedom and creative action from critical theory, the changes in family life are explained not as an effect of deinstitutionalization, but as a shift from an organized to a disorganized institutional type. This is understood as a response to changes in the wider normative structure, as a norm of individual freedom has undermined the legitimacy of the organized patriarchal nuclear family, with gender ascribed roles and associated duties. Contemporary motherhood is drawn on to illustrate the value of analysing the dynamic interactions between institutions, roles and practices for capturing both the complexity and the patterned quality of social experience.

Unlock the Pain: Exercise is the Key

Back pain is a common complaint in pregnancy, affecting approximately two-thirds of pregnant women [Pennick and Lidle, 2013]. This can lead to increased disability, affecting daily activities and cause absence from work. Evidence-based recommendations can be made for the use of exercise as an effective conservative treatment for the relief of back pain in pregnancy [Benten et al, 2014]. This poster explores the background to back pain in pregnancy and the advice women should be offered in relation to exercise to help normalise their pregnancy experience and enhance wellbeing.

Butt out - baby on board

It is estimated that the cost of treating women and their infants with smoking related problems is up to £87.5 million each year (ASH, 2013). Whilst these statistics depict a major problem for the National Health Service, they challenge midwives to become change agents within this health promotion area. A desired outcome from The Maternity Strategy for Northern Ireland (DHSSPSNI, 2012) includes giving every baby and family the best start in life. Assisting women to stop smoking before conception, could help to achieve this outcome and provide opportunities for a greater start in life following birth.

Breastfeeding: let's normalise it!

Breastfeeding rates in Northern Ireland are substantially lower in Northern Ireland compared to the rest of the United Kingdom (McAndrew et al, 2012). Midwives carry a responsibility within their health promotion activities to reiterate the benefits of breastfeeding and support women to make empowered decisions with regard to feeding their baby. This poster explores some of the challenges faced by midwives in relation to supporting women to make empowered choices and explores how midwives can help normalise breastfeeding within the clinical practice setting

Northern Ireland Infant Death Thematic Review 2015

Co-Sleeping and bed sharing are considered the social norm for approximately 90% of the world's population, with two-thirds of the world's cultures habitually practicing mother-infant co-sleeping on the same bed. Although international studies show that the practice of co-sleeping is common, it is controversial in the public health community, as many consider it a significant risk for Sudden Infant Death Syndrome (SIDS), accounting for 50% of SIDS deaths in the UK. The report offers an international review of the evidence and provides important debates and critical knowledge for both health professionals, parents and all those organisations working to support the safety of infants in their first perinatal year.

Environmental selection on transcriptome-derived SNPs in a high gene flow marine fish, the Atlantic herring (Clupea harengus)

High gene flow is considered the norm for most marine organisms and is expected to limit their ability to adapt to local environments. Few studies have directly compared the patterns of differentiation at neutral and selected gene loci in marine organisms. We analysed a transcriptome-derived panel of 281 SNPs in Atlantic herring (Clupea harengus), a highly migratory small pelagic fish, for elucidating neutral and selected genetic variation among populations and to identify candidate genes for environmental adaptation. We analysed 607 individuals from 18 spawning locations in the northeast Atlantic, including two temperature clines (5-12 °C) and two salinity clines (5-35‰). By combining genome scan and landscape genetic analyses, four genetically distinct groups of herring were identified: Baltic Sea, Baltic-North Sea transition area, North Sea/British Isles and North Atlantic; notably, samples exhibited divergent clustering patterns for neutral and selected loci. We found statistically strong evidence for divergent selection at 16 outlier loci on a global scale, and significant correlations with temperature and salinity at nine loci. On regional scales, we identified two outlier loci with parallel patterns across temperature clines and five loci associated with temperature in the North Sea/North Atlantic. Likewise, we found seven replicated outliers, of which five were significantly associated with low salinity across both salinity clines. Our results reveal a complex pattern of varying spatial genetic variation among outlier loci, likely reflecting adaptations to local environments. In addition to disclosing the fine scale of local adaptation in a highly vagile species, our data emphasize the need to preserve functionally important biodiversity.