Interpretation of association rules with multi-tier granule mining

This study was a step forward to improve the performance for discovering useful knowledge – especially, association rules in this study – in databases. The thesis proposed an approach to use granules instead of patterns to represent knowledge implicitly contained in relational databases; and multi-tier structure to interpret association rules in terms of granules. Association mappings were proposed for the construction of multi-tier structure. With these tools, association rules can be quickly assessed and meaningless association rules can be justified according to the association mappings. The experimental results indicated that the proposed approach is promising.

The relevance of shared experiences : a multi-level study of collective efficacy as a moderator of job control in the stressor-strain relationship

Karasek's Job Demand-Control model proposes that control mitigates the positive effects of work stressors on employee strain. Evidence to date remains mixed and, although a number of individual-level moderators have been examined, the role of broader, contextual, group factors has been largely overlooked. In this study, the extent to which control buffered or exacerbated the effects of demands on strain at the individual level was hypothesized to be influenced by perceptions of collective efficacy at the group level. Data from 544 employees in Australian organizations, nested within 23 workgroups, revealed significant three-way cross-level interactions among demands, control and collective efficacy on anxiety and job satisfaction. When the group perceived high levels of collective efficacy, high control buffered the negative consequences of high demands on anxiety and satisfaction. Conversely, when the group perceived low levels of collective efficacy, high control exacerbated the negative consequences of high demands on anxiety, but not satisfaction. In addition, a stress-exacerbating effect for high demands on anxiety and satisfaction was found when there was a mismatch between collective efficacy and control (i.e. combined high collective efficacy and low control). These results provide support for the notion that the stressor-strain relationship is moderated by both individual- and group-level factors.

Matching image sets via adaptive multi convex hull

Traditional nearest points methods use all the samples in an image set to construct a single convex or affine hull model for classification. However, strong artificial features and noisy data may be generated from combinations of training samples when significant intra-class variations and/or noise occur in the image set. Existing multi-model approaches extract local models by clustering each image set individually only once, with fixed clusters used for matching with various image sets. This may not be optimal for discrimination, as undesirable environmental conditions (eg. illumination and pose variations) may result in the two closest clusters representing different characteristics of an object (eg. frontal face being compared to non-frontal face). To address the above problem, we propose a novel approach to enhance nearest points based methods by integrating affine/convex hull classification with an adapted multi-model approach. We first extract multiple local convex hulls from a query image set via maximum margin clustering to diminish the artificial variations and constrain the noise in local convex hulls. We then propose adaptive reference clustering (ARC) to constrain the clustering of each gallery image set by forcing the clusters to have resemblance to the clusters in the query image set. By applying ARC, noisy clusters in the query set can be discarded. Experiments on Honda, MoBo and ETH-80 datasets show that the proposed method outperforms single model approaches and other recent techniques, such as Sparse Approximated Nearest Points, Mutual Subspace Method and Manifold Discriminant Analysis.

Physical activity recognition from accelerometer data using a multi-scale ensemble method

Accurate and detailed measurement of an individual's physical activity is a key requirement for helping researchers understand the relationship between physical activity and health. Accelerometers have become the method of choice for measuring physical activity due to their small size, low cost, convenience and their ability to provide objective information about physical activity. However, interpreting accelerometer data once it has been collected can be challenging. In this work, we applied machine learning algorithms to the task of physical activity recognition from triaxial accelerometer data. We employed a simple but effective approach of dividing the accelerometer data into short non-overlapping windows, converting each window into a feature vector, and treating each feature vector as an i.i.d training instance for a supervised learning algorithm. In addition, we improved on this simple approach with a multi-scale ensemble method that did not need to commit to a single window size and was able to leverage the fact that physical activities produced time series with repetitive patterns and discriminative features for physical activity occurred at different temporal scales.

Multi-shot person re-identification via relational stein divergence

Person re-identification is particularly challenging due to significant appearance changes across separate camera views. In order to re-identify people, a representative human signature should effectively handle differences in illumination, pose and camera parameters. While general appearance-based methods are modelled in Euclidean spaces, it has been argued that some applications in image and video analysis are better modelled via non-Euclidean manifold geometry. To this end, recent approaches represent images as covariance matrices, and interpret such matrices as points on Riemannian manifolds. As direct classification on such manifolds can be difficult, in this paper we propose to represent each manifold point as a vector of similarities to class representers, via a recently introduced form of Bregman matrix divergence known as the Stein divergence. This is followed by using a discriminative mapping of similarity vectors for final classification. The use of similarity vectors is in contrast to the traditional approach of embedding manifolds into tangent spaces, which can suffer from representing the manifold structure inaccurately. Comparative evaluations on benchmark ETHZ and iLIDS datasets for the person re-identification task show that the proposed approach obtains better performance than recent techniques such as Histogram Plus Epitome, Partial Least Squares, and Symmetry-Driven Accumulation of Local Features.

Interpretation of association rules in multi-tier structures

Dealing with the large amount of data resulting from association rule mining is a big challenge. The essential issue is how to provide efficient methods for summarizing and representing meaningful discovered knowledge from databases. This paper presents a new approach called multi-tier granule mining to improve the performance of association rule mining. Rather than using patterns, it uses granules to represent knowledge that is implicitly contained in relational databases. This approach also uses multi-tier structures and association mappings to interpret association rules in terms of granules. Consequently, association rules can be quickly assessed and meaningless association rules can be justified according to these association mappings. The experimental results indicate that the proposed approach is promising

Unidentified vehicle in multi-vehicle accident - failure to notify Nominal Defendant no bar to proceedings against it

In Kumar v Suncorp Metway Insurance Limited [2004] QSC 381 Douglas J examined s37 of the Motor Accident Insurance Act 1994 (Qld) in the context of an accident involving multiple insurers when a notice of accident had not been given to the Nominal Defendant

Barriers to implementing eHealth : a multi-dimensional perspective

This paper strives to identify barriers that hamper eHealth implementation from different perspectives. The benefits offered by eHealth and the need for eHealth preparedness is first discussed. This is followed by a discussion on the integral components of a robust eHealth infrastructure. Then, the barriers to eHealth such as technical interoperability issues, lack of holistic approach and technology disconnect are explained in detail. Finally, solutions to promote better adoption of eHealth through government policies, standardisation and training are also discussed.

Multi-utility finance : a problematic case

This paper treats one particular version of the multi-utility strategy as experienced by the Hyder Group. We examine some aspectw of the company's financial performance and consider the implications.

Multi-scale bio-inspired place recognition

This paper presents a novel place recognition algorithm inspired by the recent discovery of overlapping and multi-scale spatial maps in the rodent brain. We mimic this hierarchical framework by training arrays of Support Vector Machines to recognize places at multiple spatial scales. Place match hypotheses are then cross-validated across all spatial scales, a process which combines the spatial specificity of the finest spatial map with the consensus provided by broader mapping scales. Experiments on three real-world datasets including a large robotics benchmark demonstrate that mapping over multiple scales uniformly improves place recognition performance over a single scale approach without sacrificing localization accuracy. We present analysis that illustrates how matching over multiple scales leads to better place recognition performance and discuss several promising areas for future investigation.

Amplitude equations and asymptotic expansions for multi-scale problems

In this paper we introduce a new technique to obtain the slow-motion dynamics in nonequilibrium and singularly perturbed problems characterized by multiple scales. Our method is based on a straightforward asymptotic reduction of the order of the governing differential equation and leads to amplitude equations that describe the slowly-varying envelope variation of a uniformly valid asymptotic expansion. This may constitute a simpler and in certain cases a more general approach toward the derivation of asymptotic expansions, compared to other mainstream methods such as the method of Multiple Scales or Matched Asymptotic expansions because of its relation with the Renormalization Group. We illustrate our method with a number of singularly perturbed problems for ordinary and partial differential equations and recover certain results from the literature as special cases. © 2010 - IOS Press and the authors. All rights reserved.

Improvement of a dynamic accumulator at ICICS 07 and its application in multi-user keyword-based retrieval on encrypted data

A dynamic accumulator is an algorithm, which merges a large set of elements into a constant-size value such that for an element accumulated, there is a witness confirming that the element was included into the value, with a property that accumulated elements can be dynamically added and deleted into/from the original set. Recently Wang et al. presented a dynamic accumulator for batch updates at ICICS 2007. However, their construction suffers from two serious problems. We analyze them and propose a way to repair their scheme. We use the accumulator to construct a new scheme for common secure indices with conjunctive keyword-based retrieval.

Large networks of vertical multi-layer graphenes with morphology-tunable magnetoresistance

We report on the comparative study of magnetotransport properties of large-area vertical few-layer graphene networks with different morphologies, measured in a strong (up to 10 T) magnetic field over a wide temperature range. The petal-like and tree-like graphene networks grown by a plasma enhanced CVD process on a thin (500 nm) silicon oxide layer supported by a silicon wafer demonstrate a significant difference in the resistance-magnetic field dependencies at temperatures ranging from 2 to 200 K. This behaviour is explained in terms of the effect of electron scattering at ultra-long reactive edges and ultra-dense boundaries of the graphene nanowalls. Our results pave a way towards three-dimensional vertical graphene-based magnetoelectronic nanodevices with morphology-tuneable anisotropic magnetic properties. © The Royal Society of Chemistry 2013.

Advanced mass spectrometry-based multi-omics technologies for exploring the pathogenesis of hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is one of the primary hepatic malignancies and is the third most common cause of cancer related death worldwide. Although a wealth of knowledge has been gained concerning the initiation and progression of HCC over the last half century, efforts to improve our understanding of its pathogenesis at a molecular level are still greatly needed, to enable clinicians to enhance the standards of the current diagnosis and treatment of HCC. In the post-genome era, advanced mass spectrometry driven multi-omics technologies (e.g., profiling of DNA damage adducts, RNA modification profiling, proteomics, and metabolomics) stand at the interface between chemistry and biology, and have yielded valuable outcomes from the study of a diversity of complicated diseases. Particularly, these technologies are being broadly used to dissect various biological aspects of HCC with the purpose of biomarker discovery, interrogating pathogenesis as well as for therapeutic discovery. This proof of knowledge-based critical review aims at exploring the selected applications of those defined omics technologies in the HCC niche with an emphasis on translational applications driven by advanced mass spectrometry, toward the specific clinical use for HCC patients. This approach will enable the biomedical community, through both basic research and the clinical sciences, to enhance the applicability of mass spectrometry-based omics technologies in dissecting the pathogenesis of HCC and could lead to novel therapeutic discoveries for HCC.