Representation of analysis results involving aleatory and epistemic uncertainty

Procedures are described for the representation of results in analyses that involve both aleatory uncertainty and epistemic uncertainty, with aleatory uncertainty deriving from an inherent randomness in the behaviour of the system under study and epistemic uncertainty deriving from a lack of knowledge about the appropriate values to use for quantities that are assumed to have fixed but poorly known values in the context of a specific study. Aleatory uncertainty is usually represented with probability and leads to cumulative distribution functions (CDFs) or complementary CDFs (CCDFs) for analysis results of interest. Several mathematical structures are available for the representation of epistemic uncertainty, including interval analysis, possibility theory, evidence theory and probability theory. In the presence of epistemic uncertainty, there is not a single CDF or CCDF for a given analysis result. Rather, there is a family of CDFs and a corresponding family of CCDFs that derive from epistemic uncertainty and have an uncertainty structure that derives from the particular uncertainty structure (e.g. interval analysis, possibility theory, evidence theory or probability theory) used to represent epistemic uncertainty. Graphical formats for the representation of epistemic uncertainty in families of CDFs and CCDFs are investigated and presented for the indicated characterisations of epistemic uncertainty.     

Characteristics of college cyberbullies

Cyberbullying is a newer variation of bullying that utilizes technology to harass. This study investigated the psychological symptomology, suicidal behaviors, aggressive tendencies, and illegal behaviors of college cyberbullies. Sixty cyberbullies and 19 cyberbully/victims (participants who cyberbullied and were victims of cyberbullying) scored higher in psychological symptoms of depression, hostility, interpersonal sensitivity, paranoia, phobic anxiety, and psychoticism than participants who were uninvolved with cyberbullying. The cyberbullies also endorsed more suicidal behaviors and scored higher in aggression. The cyberbully/victims engaged in more violent and drug crimes than cyberbullies and controls. These findings indicate perpetrators of cyberbullying have more psychological distress, aggressive tendencies, and engage in more illegal behaviors than those who do not cyberbully. This is the first study to show the distress cyberbullies are experiencing.     

Personalising speech-to-speech translation: Unsupervised cross-lingual speaker adaptation for HMM-based speech synthesis

In this paper we present results of unsupervised cross-lingual speaker adaptation applied to text-to-speech synthesis. The application of our research is the personalisation of speech-to-speech translation in which we employ a HMM statistical framework for both speech recognition and synthesis. This framework provides a logical mechanism to adapt synthesised speech output to the voice of the user by way of speech recognition. In this work we present results of several different unsupervised and cross-lingual adaptation approaches as well as an end-to-end speaker adaptive speech-to-speech translation system. Our experiments show that we can successfully apply speaker adaptation in both unsupervised and cross-lingual scenarios and our proposed algorithms seem to generalise well for several language pairs. We also discuss important future directions including the need for better evaluation metrics.     

Network numerical analysis for the smoother and the lagged joint-process estimator

Motivated by the fact that the a priori least-squares-order-recursive lattice (LSORL) smoother is more robust than the LSORL joint-process estimator with lagged desired signals in the finite precision, we model numerical properties of the two algorithms by virtue of previous efforts. Then, we give the reason why the smoother is substantially more robust than the lagged joint-process estimator by providing the explicit analysis for the performance difference of the two algorithms.     

Growth and feeding efficiency of wild and aquaculture genotypes of rainbow trout (Oncorhynchus mykiss) common to Lake Huron,Canada

Selective breeding of salmonid fishes for the purpose of commercial aquaculture has resulted in domesticated strains possessing a divergent physiological and behavioral phenotype from that of wild conspecifics. Freshwater production of rainbow trout (Oncorhynchus mykiss) has been occurring in regions of Lake Huron, Canada, for decades yet the growth and performance of domestic (aquaculture) versus wild (naturalized) strains are poorly understood. We conducted two trials to examine growth differences between size-matched wild and domestic strains of juvenile rainbow trout: (1) reared separately and fed to satiation; and (2) reared together and fed a reduced ration to induce competition. Additionally, we used bioenergetics models to assess strain-specific growth rates across a range of water temperatures (5, 10, 15 and 18 °C) as well as Lake Huron temperatures during the open-water season. Domestic rainbow trout showed a growth advantage throughout the 102 d trials, and by the end of the study had achieved a mass > two-fold that of the wild strain and had greater fork length, condition, and thermal growth coefficient (TGC) under both treatments. Rapid growth of domestic strain fish was achieved through the combination of enhanced feed consumption (by ~ 40%) and feeding efficiency (up to 60% lower feed conversion ratio) relative to wild fish. Divergence in growth rates between strains was most pronounced (> 3 ×) when modeled with Lake Huron open-water temperatures. We demonstrate that the growth and feed-conversion efficiency differ significantly between these two strains of rainbow trout under laboratory conditions, suggesting that differences could be even greater in nature.     

Complete Boolean Satisfiability Solving Algorithms Based on Local Search

Boolean satisfiability (SAT) is a well-known problem in computer science, artificial intelligence, and operations research. This paper focuses on the satisfiability problem of Model RB structure that is similar to graph coloring problems and others. We propose a translation method and three effective complete SAT solving algorithms based on the characterization of Model RB structure. We translate clauses into a graph with exclusive sets and relative sets. In order to reduce search depth, we determine search order using vertex weights and clique in the graph. The results show that our algorithms are much more effective than the best SAT solvers in numerous Model RB benchmarks, especially in those large benchmark instances.     

FraRI: an algorithm to account for the discontinuity at 60° N in MODIS temperature products

The Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA's Terra and Aqua satellites conducts continuous monitoring of the Earth's land surface and oceans. Recently, a sharp discontinuity (averaging 1.9°C) has been noticed at 60° N in both MODIS daytime and night-time land surface temperature (LST) products. This linear artefact arises because the CO₂ high cloud test in the operational code for the generation of the MODIS cloud mask product is used only in the non-polar region (between 60° N and 60° S). The resulting discontinuity clearly has negative implications for any statistical applications of these temperature data. In this technical note we present a new algorithm, which minimizes this discontinuity. The method uses edge detection and elimination based on a mixture of Sobel and non-linear Laplacian filters (edge detection and quantification), cubic splines (edge modelling), and a controllable power function for image restoration. The implementation of this algorithm is demonstrated on an image of average minimum night-time LST between 2001 and 2008.     

Winter and spring surface velocity fields in the Cape Blanc region as deduced with the maximum cross-correlation technique

The ocean surface velocity field in the Cape Blanc region, off Northwest Africa, is investigated with the maximum cross-correlation (MCC) method applied to channel-4 Advanced Very High Resolution Radiometer satellite images. An initial sensitivity analysis allows us to select the four parameters that provide maximum area coverage and the best velocity resolution, while limiting the standard deviation for each velocity component within reasonable values. These are (m, n, MV, CT)?=?(22, 32, 50, 0.6), where m and n are the number of pixels of the search window (SW) and reference window (RW), respectively, MV is the maximum possible velocity (in cm s^?1), and CT is a correlation threshold for a feature to be tracked. A total of 489 images, for years 2005 and 2006, are analysed, and 106 velocity maps are generated with good coverage of the coastal transition zone (CTZ), most of them for the winter (34) and spring (59) seasons. We remove spurious data using the method's own filters (MV, CT, and a neighbour-vector comparison), requesting the velocity components to have Gaussian distributions and smoothing the resulting velocity fields with a median-vector filter. The instantaneous velocity maps illustrate the response of the alongshore coastal jet north of Cape Blanc (and its extension along the Cape Verde frontal region) to wind forcing, as well as the presence of numerous mesoscalar features (100–300 km wide) superposed on a westward offshore transport south of Cape Blanc. We also produce mean and standard deviation winter and spring velocity maps, which are compared with the corresponding mean sea surface temperature fields. The along-shore and offshore flow is better defined and is more intense in spring than in winter, in concordance with cross-slope sharper temperature gradients during this season, and brings about a cooling of the whole region. We identify five different ubiquitous currents: a southwestward jet north of Cape Blanc, a northwestward jet off Banc d'Argin, an offshore convergent jet, a spring jet-like feature at 18° N, and a southward flow in the southwestern CTZ.     

Product-forms in batch networks: Approximation and asymptotics

It is shown that a Markovian queue, with bulk arrivals and departures having any probability mass functions for their batch sizes, has geometrically distributed queue length at equilibrium (when this exists) provided there is an additional special bulk arrival stream, with particular rate and batch size distribution, when the server is idle. It is shown that the time-averaged input rate of the special arrivals tends to zero as the queue becomes saturated, and a heavy-traffic limit for the queue without special arrivals is derived by martingale methods. This is shown to give the same asymptotic queue length probabilities as the geometric model. The product form is then extended to tandem networks of batch queues using the reversed compound agent theorem (RCAT). In order to obtain the product form in this case, it is required that, in addition to special arrival streams, so-called ‘partial batches’ are discarded immediately from the network when there are not enough customers in the queue to fill an entire departing batch. Somewhat surprisingly it turns out that, in heavy traffic, the product-form network does not always agree with the regulated Brownian motion (RBM) diffusion limit for the standard network without special arrivals and where partial batches are not discarded, but forwarded to the next node. Indeed, we show that the two models agree in heavy traffic if and only if the skew-symmetry condition for the RBM to have a product form is satisfied. When the condition does hold, our theoretical and numerical results thus validate the use of the product-form batch networks as moderate-traffic approximations to the analogous standard queueing network model without special arrivals and where partial batches may be forwarded to the next node instead of being lost. In the case that the condition does not hold, we obtain a new product-form stationary distribution for the associated non-RBM diffusion limit.