An orthogonal forward regression technique for sparse kernel density estimation

Using the classical Parzen window (PW) estimate as the desired response, the kernel density estimation is formulated as a regression problem and the orthogonal forward regression technique is adopted to construct sparse kernel density (SKD) estimates. The proposed algorithm incrementally minimises a leave-one-out test score to select a sparse kernel model, and a local regularisation method is incorporated into the density construction process to further enforce sparsity. The kernel weights of the selected sparse model are finally updated using the multiplicative nonnegative quadratic programming algorithm, which ensures the nonnegative and unity constraints for the kernel weights and has the desired ability to reduce the model size further. Except for the kernel width, the proposed method has no other parameters that need tuning, and the user is not required to specify any additional criterion to terminate the density construction procedure. Several examples demonstrate the ability of this simple regression-based approach to effectively construct a SKID estimate with comparable accuracy to that of the full-sample optimised PW density estimate. (c) 2007 Elsevier B.V. All rights reserved.

Land-cover classification from airborne LIDAR data fused with aerial optical images

Airborne LIght Detection And Ranging (LIDAR) provides accurate height information for objects on the earth, which makes LIDAR become more and more popular in terrain and land surveying. In particular, LIDAR data offer vital and significant features for land-cover classification which is an important task in many application domains. In this paper, an unsupervised approach based on an improved fuzzy Markov random field (FMRF) model is developed, by which the LIDAR data, its co-registered images acquired by optical sensors, i.e. aerial color image and near infrared image, and other derived features are fused effectively to improve the ability of the LIDAR system for the accurate land-cover classification. In the proposed FMRF model-based approach, the spatial contextual information is applied by modeling the image as a Markov random field (MRF), with which the fuzzy logic is introduced simultaneously to reduce the errors caused by the hard classification. Moreover, a Lagrange-Multiplier (LM) algorithm is employed to calculate a maximum A posteriori (MAP) estimate for the classification. The experimental results have proved that fusing the height data and optical images is particularly suited for the land-cover classification. The proposed approach works very well for the classification from airborne LIDAR data fused with its coregistered optical images and the average accuracy is improved to 88.9%.

Dynamics of brand competition: Effects of unobserved social networks

Brand competition is modelled using an agent based approach in order to examine the long run dynamics of market structure and brand characteristics. A repeated game is designed where myopic firms choose strategies based on beliefs about their rivals and consumers. Consumers are heterogeneous and can observe neighbour behaviour through social networks. Although firms do not observe them, the social networks have a significant impact on the emerging market structure. Presence of networks tends to polarize market share and leads to higher volatility in brands. Yet convergence in brand characteristics usually happens whenever the market reaches a steady state. Scale-free networks accentuate the polarization and volatility more than small world or random networks. Unilateral innovations are less frequent under social networks.

Handling single node failures using agents in computer clusters

The work reported in this paper is motivated towards handling single node failures for parallel summation algorithms in computer clusters. An agent based approach is proposed in which a task to be executed is decomposed to sub-tasks and mapped onto agents that traverse computing nodes. The agents intercommunicate across computing nodes to share information during the event of a predicted node failure. Two single node failure scenarios are considered. The Message Passing Interface is employed for implementing the proposed approach. Quantitative results obtained from experiments reveal that the agent based approach can handle failures more efficiently than traditional failure handling approaches.

An RCT study to evaluate a targeted, theory driven healthy eating leaflet

A theory based healthy eating leaflet was evaluated against an existing publicly available standard leaflet. The intervention leaflet was designed to encourage healthy eating in 18-30 year olds and was developed by modifying an existing British Nutrition Foundation leaflet. The intervention leaflet targeted attitudes and self-efficacy. Participants (n=104) were randomly assigned either to the intervention, Foundation or a local food leaflet control condition. Cognitions were measured pre-intervention, immediately after reading the corresponding leaflet, and once again at two weeks follow-up. Critically, intentions to eat healthily were significantly greater at follow-up in the Intervention group compared to the other two groups, with the former leaflet also being perceived as more persuasive. The Intervention group also showed evidence of healthier eating at two weeks compared to the other two groups. Collectively the results illustrate the utility of a targeted theory-based approach.

Evaluation of peak-picking algorithms for protein mass spectrometry

Peak picking is an early key step in MS data analysis. We compare three commonly used approaches to peak picking and discuss their merits by means of statistical analysis. Methods investigated encompass signal-to-noise ratio, continuous wavelet transform, and a correlation-based approach using a Gaussian template. Functionality of the three methods is illustrated and discussed in a practical context using a mass spectral data set created with MALDI-TOF technology. Sensitivity and specificity are investigated using a manually defined reference set of peaks. As an additional criterion, the robustness of the three methods is assessed by a perturbation analysis and illustrated using ROC curves.

Untitled (Abstraction of labour time eternal recurrence monad)

An AHRC funded project titled: Picturing ideas? Visualising and Synthesising Ideas as art (2009-10). Outputs including: 4 exhibitions; 4 publications; 3 papers; 2 largescale backlit digital prints; 1 commissioned print. (See Additional Information) ----ABSTRACT: Utilising the virtuality of digital imagery this practice-led project explored the possibility of the cross-articulation between text and image and the bridging or synthesising potential of the visual affect of ideas. A series of digital images were produced 'picturing' or 'visualising' philosophical ideas derived from the writings of the philosopher Giles Deleuze, as remodellings of pre-existing philosophical ideas; developed through dialogues and consultation with specialists in the fields from which the ideas were drawn (philosophy, psychology, film) as well as artists and theorists concerned with ideas of 'mental imagery' and visualisation. Final images were produced as a synthesis (or combination) of these visualisations and presented in the format of large scale, backlit digital prints at a series of prestigious international exhibitions (see details above). Evaluation took the form of a four page illustrated text in Frieze magazine (August 2009) and three papers delivered at University of Ulster, Goldsmiths College of Art and Loughborough University. The project also included the publication of a catalogue essay (EAST 09) and an illustrated poem (in the Dark Monarch publication). A print version of the image was commissioned by Invisible Exports Gallery, New York and subsequently exhibited in The Devos Art Museum, School of Art & Design at Northern Michigan University and in a publication edited by Cedar Lewisohn for Tate Publishing. The project was funded by an AHRC practice-led grant (17K) and Arts Council of England award (1.5K). The outputs, including high profile, publicly accessible exhibitions, prestigious publications and conference papers ensured the dissemination of the research to a wide range of audiences, including scholars/researchers across the arts and humanities engaged in practice-based and interdisciplinary theoretical work (in particular in the fields of contemporary art and art theory and those working on the integration of art and theory/philosophy/psychology) but also the wider audience for contemporary art.

Panzootics and the poor: devising a global livestock disease prioritisation framework for poverty alleviation

Panzootics such as highly pathogenic avian influenza and Rift Valley fever have originated from the South, largely among poor communities. On a global level, approximately two-thirds of those individuals living on less than US$2 per day keep livestock. Consequently, there is a need to better target animal health interventions for poverty reduction using an evidence-based approach. Therefore, the paper offers a three-step prioritisation framework using calculations derived from standard poverty measures: the poverty gap and the head count ratio. Data from 265 poor livestock-keeping households in Kenya informed the study. The results demonstrate that, across a spectrum of producers, the dependence upon particular species varies. Furthermore, the same livestock disease has differing impacts on the depth and severity of poverty. Consequently, animal health interventions need to

Symmetric Push-Sum Protocol for decentralised aggregation

Gossip (or Epidemic) protocols have emerged as a communication and computation paradigm for large-scale networked systems. These protocols are based on randomised communication, which provides probabilistic guarantees on convergence speed and accuracy. They also provide robustness, scalability, computational and communication efficiency and high stability under disruption. This work presents a novel Gossip protocol named Symmetric Push-Sum Protocol for the computation of global aggregates (e.g., average) in decentralised and asynchronous systems. The proposed approach combines the simplicity of the push-based approach and the efficiency of the push-pull schemes. The push-pull schemes cannot be directly employed in asynchronous systems as they require synchronous paired communication operations to guarantee their accuracy. Although push schemes guarantee accuracy even with asynchronous communication, they suffer from a slower and unstable convergence. Symmetric Push- Sum Protocol does not require synchronous communication and achieves a convergence speed similar to the push-pull schemes, while keeping the accuracy stability of the push scheme. In the experimental analysis, we focus on computing the global average as an important class of node aggregation problems. The results have confirmed that the proposed method inherits the advantages of both other schemes and outperforms well-known state of the art protocols for decentralized Gossip-based aggregation.

The North Atlantic jet stream under climate change and its relation to the NAO and EA patterns

This paper describes recent variations of the North Atlantic eddy-driven jet stream and analyzes the mean response of the jet to anthropogenic forcing in climate models. Jet stream changes are analyzed both using a direct measure of the near-surface westerly wind maximum and using an EOF-based approach. This allows jet stream changes to be related to the widely used leading patterns of variability: the North Atlantic Oscillation (NAO) and East Atlantic (EA) pattern. Viewed in NAO–EA state space, isolines of jet latitude and speed resemble a distorted polar coordinate system, highlighting the dependence of the jet stream quantities on both spatial patterns. Some differences in the results of the two methods are discussed, but both approaches agree on the general characteristics of the climate models. While there is some agreement between models on a poleward shift of the jet stream in response to anthropogenic forcing, there is still considerable spread between different model projections, especially in winter. Furthermore, the model responses to forcing are often weaker than their biases when compared to a reanalysis. Diagnoses of jet stream changes can be sensitive to the methodologies used, and several aspects of this are also discussed.

Parallel Architectures for High Speed Multipliers

The authors compare various array multiplier architectures based on (p,q) counter circuits. The tradeoff in multiplier design is always between adding complexity and increasing speed. It is shown that by using a (2,2,3) counter cell it is possible to gain a significant increase in speed over a conventional full-adder, carry-save array based approach. The increase in complexity should be easily accommodated using modern emitter-coupled-logic processes.

Understanding the long term success of UK construction firms: the extent and role of 'hidden' corporate social responsibility

The construction sector is often described as lagging behind other major industries. At first this appears fair when considering the concept of corporate social responsibility (CSR). It is argued that CSR is ill-defined, with firms struggling to make sense of and engage with it. Literature suggests that the short-termism view of construction firms renders the long-term, triple-bottom-line principle of CSR untenable. This seems to be borne out by literature indicating that construction firms typically adopt a compliance-based approach to CSR instead of discretionary CSR which is regarded as adding most value to firms and benefiting the broadest group of stakeholders. However, this research conducted in the UK using a regional construction firm offers a counter argument whereby discretionary CSR approaches are well embedded and enacted within the firms’ business operations even though they are not formally articulated as CSR strategies and thus remain 'hidden'. This raises questions in the current CSR debate. First, is ‘hidden’ CSR relevant to the long term success of construction firms? and to what extent do these firms need to reinvent themselves to formally take advantage of the CSR agenda?

Automatic keyphrase extraction: a comparison of methods

There are many published methods available for creating keyphrases for documents. Previous work in the field has shown that in a significant proportion of cases author selected keyphrases are not appropriate for the document they accompany. This requires the use of such automated methods to improve the use of keyphrases. Often the keyphrases are not updated when the focus of a paper changes or include keyphrases that are more classificatory than explanatory. The published methods are all evaluated using different corpora, typically one relevant to their field of study. This not only makes it difficult to incorporate the useful elements of algorithms in future work but also makes comparing the results of each method inefficient and ineffective. This paper describes the work undertaken to compare five methods across a common baseline of six corpora. The methods chosen were term frequency, inverse document frequency, the C-Value, the NC-Value, and a synonym based approach. These methods were compared to evaluate performance and quality of results, and to provide a future benchmark. It is shown that, with the comparison metric used for this study Term Frequency and Inverse Document Frequency were the best algorithms, with the synonym based approach following them. Further work in the area is required to determine an appropriate (or more appropriate) comparison metric.

Virulotyping and antimicrobial resistance typing of salmonella enterica serovars relevant to human health in Europe

The combination of virulence gene and antimicrobial resistance gene typing using DNA arrays is a recently developed genomics-based approach to bacterial molecular epidemiology. We have now applied this technology to 523 Salmonella enterica subsp. enterica strains collected from various host sources and public health and veterinary institutes across nine European countries. The strain set included the five predominant Salmonella serovars isolated in Europe (Enteritidis, Typhimurium, Infantis, Virchow, and Hadar). Initially, these strains were screened for 10 potential virulence factors (avrA, ssaQ, mgtC, siiD, sopB, gipA, sodC1, sopE1, spvC, and bcfC) by polymerase chain reaction. The results indicated that only 14 profiles comprising these genes (virulotypes) were observed throughout Europe. Moreover, most of these virulotypes were restricted to only one (n = 9) or two (n = 4) serovars. The data also indicated that the virulotype did not vary significantly with host source or geographical location. Subsequently, a representative subset of 77 strains was investigated using a microarray designed to detect 102 virulence and 49 resistance determinants. The results confirmed and extended the previous observations using the virulo-polymerase chain reaction screen. Strains belonging to the same serovar grouped together, indicating that the broader virulence-associated gene complement corresponded with the serovar. There were, however, some differences in the virulence gene profiles between strains belonging to an individual serovar. This variation occurred primarily within those virulence genes that were prophage encoded, in fimbrial clusters or in the virulence plasmid. It seems likely that such changes enable Salmonella to adapt to different environmental conditions, which might be reflected in serovar-specific ecology. In this strain subset a number of resistance genes were detected and were serovar restricted to a varying degree. Once again the profiles of those genes encoding resistance were similar or the same for each serovar in all hosts and countries investigated.