Structure of glassy GeO2

The full set of partial structure factors for glassy germania, or GeO2, were accurately measured by using the method of isotopic substitution in neutron diffraction in order to elucidate the nature of the pair correlations for this archetypal strong glass former. The results show that the basic tetrahedral Ge(O-1/2)(4) building blocks share corners with a mean inter-tetrahedral Ge-O-Ge bond angle of 132(2)degrees. The topological and chemical ordering in the resultant network displays two characteristic length scales at distances greater than the nearest neighbour. One of these describes the intermediate range order, and manifests itself by the appearance of a first sharp diffraction peak in the measured diffraction patterns at a scattering vector k(FSDP) approximate to 1.53 angstrom(-1), while the other describes so-called extended range order, and is associated with the principal peak at k(PP) = 2.66( 1) angstrom(-1). We find that there is an interplay between the relative importance of the ordering on these length scales for tetrahedral network forming glasses that is dominated by the extended range ordering with increasing glass fragility. The measured partial structure factors for glassy GeO2 are used to reproduce the total structure factor measured by using high energy x-ray diffraction and the experimental results are also compared to those obtained by using classical and first principles molecular dynamics simulations.

Topological versus chemical ordering in network glasses at intermediate and extended length scales

Atomic ordering in network glasses on length scales longer than nearest-neighbour length scales has long been a source of controversy(1-6). Detailed experimental information is therefore necessary to understand both the network properties and the fundamentals of glass formation. Here we address the problem by investigating topological and chemical ordering in structurally disordered AX2 systems by applying the method of isotopic substitution in neutron diffraction to glassy ZnCl2. This system may be regarded as a prototypical ionic network forming glass, provided that ion polarization effects are taken into account(7), and has thus been the focus of much attention(8-14). By experiment, we show that both the topological and chemical ordering are described by two length scales at distances greater than nearest-neighbour length scales. One of these is associated with the intermediate range, as manifested by the appearance in the measured diffraction patterns of a first sharp diffraction peak at 1.09( 3) angstrom(-1); the other is associated with an extended range, which shows ordering in the glass out to 62( 4) angstrom. We also find that these general features are characteristic of glassy GeSe2, a prototypical covalently bonded network material(15,16). The results therefore offer structural insight into those length scales that determine many important aspects of supercooled liquid and glass phenomenology(11).

Visualisation of bioinformatics datasets

Analysing the molecular polymorphism and interactions of DNA, RNA and proteins is of fundamental importance in biology. Predicting functions of polymorphic molecules is important in order to design more effective medicines. Analysing major histocompatibility complex (MHC) polymorphism is important for mate choice, epitope-based vaccine design and transplantation rejection etc. Most of the existing exploratory approaches cannot analyse these datasets because of the large number of molecules with a high number of descriptors per molecule. This thesis develops novel methods for data projection in order to explore high dimensional biological dataset by visualising them in a low-dimensional space. With increasing dimensionality, some existing data visualisation methods such as generative topographic mapping (GTM) become computationally intractable. We propose variants of these methods, where we use log-transformations at certain steps of expectation maximisation (EM) based parameter learning process, to make them tractable for high-dimensional datasets. We demonstrate these proposed variants both for synthetic and electrostatic potential dataset of MHC class-I. We also propose to extend a latent trait model (LTM), suitable for visualising high dimensional discrete data, to simultaneously estimate feature saliency as an integrated part of the parameter learning process of a visualisation model. This LTM variant not only gives better visualisation by modifying the project map based on feature relevance, but also helps users to assess the significance of each feature. Another problem which is not addressed much in the literature is the visualisation of mixed-type data. We propose to combine GTM and LTM in a principled way where appropriate noise models are used for each type of data in order to visualise mixed-type data in a single plot. We call this model a generalised GTM (GGTM). We also propose to extend GGTM model to estimate feature saliencies while training a visualisation model and this is called GGTM with feature saliency (GGTM-FS). We demonstrate effectiveness of these proposed models both for synthetic and real datasets. We evaluate visualisation quality using quality metrics such as distance distortion measure and rank based measures: trustworthiness, continuity, mean relative rank errors with respect to data space and latent space. In cases where the labels are known we also use quality metrics of KL divergence and nearest neighbour classifications error in order to determine the separation between classes. We demonstrate the efficacy of these proposed models both for synthetic and real biological datasets with a main focus on the MHC class-I dataset.

Implicit Feature Selection with the Value Difference Metric

Publisher PDF

A tight-binding analysis of the electronic properties of III-nitride semiconductors

This thesis divides into two distinct parts, both of which are underpinned by the tight-binding model. The first part covers our implementation of the tight-binding model in conjunction with the Berry phase theory of electronic polarisation to probe the atomistic origins of spontaneous polarisation and piezoelectricity as well as attempting to accurately calculate the values and coefficients associated with these phenomena. We first develop an analytic model for the polarisation of a one-dimensional linear chain of atoms. We compare the zincblende and ideal wurtzite structures in terms of effective charges, spontaneous polarisation and piezoelectric coefficients, within a first nearest neighbour tight-binding model. We further compare these to real wurtzite structures and conclude that accurate quantitative results are beyond the scope of this model but qualitative trends can still be described. The second part of this thesis deals with implementing the tight-binding model to investigate the effect of local alloy fluctuations in bulk AlGaN alloys and InGaN quantum wells. We calculate the band gap evolution of Al1_xGaxN across the full composition range and compare it to experiment as well as fitting bowing parameters to the band gap as well as to the conduction band and valence band edges. We also investigate the wavefunction character of the valence band edge to determine the composition at which the optical polarisation switches in Al1_xGaxN alloys. Finally, we examine electron and hole localisation in InGaN quantum wells. We show how the built-in field localises the carriers along the c-axis and how local alloy fluctuations strongly localise the highest hole states in the c-plane, while the electrons remain delocalised in the c-plane. We show how this localisation affects the charge density overlap and also investigate the effect of well width fluctuations on the localisation of the electrons.

Performance Evaluation of a Statistical and a Neural Network Model for Nonrigid Shape-Based Registration

Shape-based registration methods frequently encounters in the domains of computer vision, image processing and medical imaging. The registration problem is to find an optimal transformation/mapping between sets of rigid or nonrigid objects and to automatically solve for correspondences. In this paper we present a comparison of two different probabilistic methods, the entropy and the growing neural gas network (GNG), as general feature-based registration algorithms. Using entropy shape modelling is performed by connecting the point sets with the highest probability of curvature information, while with GNG the points sets are connected using nearest-neighbour relationships derived from competitive hebbian learning. In order to compare performances we use different levels of shape deformation starting with a simple shape 2D MRI brain ventricles and moving to more complicated shapes like hands. Results both quantitatively and qualitatively are given for both sets.

Rotina computacional de obtenção, tratamento e visualização de dados oceanográficos

Atualmente, sensores remotos e computadores de alto desempenho estão sendo utilizados como instrumentos principais na coleta e produção de dados oceanográficos. De posse destes dados, é possível realizar estudos que permitem simular e prever o comportamento do oceano por meio de modelos numéricos regionais. Dentre os fatores importantes no estudo da oceanografia, podem ser destacados àqueles referentes aos impactos ambientais, de contaminação antrópica, utilização de energias renováveis, operações portuárias e etc. Contudo, devido ao grande volume de dados gerados por instituições ambientais, na forma de resultados de modelos globais como o HYCOM (Hybrid Coordinate Ocean Model) e dos programas de Reanalysis da NOAA (National Oceanic and Atmospheric Administration), torna-se necessária a criação de rotinas computacionais para realizar o tratamento de condições iniciais e de contorno, de modo que possam ser aplicadas a modelos regionais como o TELEMAC3D (www.opentelemac.org). Problemas relacionados a baixa resolução, ausência de dados e a necessidade de interpolação para diferentes malhas ou sistemas de coordenadas verticais, tornam necessária a criação de um mecanismo computacional que realize este tratamento adequadamente. Com isto, foram desenvolvidas rotinas na linguagem de programação Python, empregando interpoladores de vizinho mais próximo, de modo que, a partir de dados brutos dos modelos HYCOM e do programa de Reanalysis da NOAA, foram preparadas condições iniciais e de contorno para a realização de uma simulação numérica teste. Estes resultados foram confrontados com outro resultado numérico onde, as condições foram construídas a partir de um método de interpolação mais sofisticado, escrita em outra linguagem, e que já vem sendo utilizada no laboratório. A análise dos resultados permitiu concluir que, a rotina desenvolvida no âmbito deste trabalho, funciona adequadamente para a geração de condições iniciais e de contorno do modelo TELEMAC3D. Entretanto, um interpolador mais sofisticado deve ser desenvolvido de forma a aumentar a qualidade nas interpolações, otimizar o custo computacional, e produzir condições que sejam mais realísticas para a utilização do modelo TELEMAC3D.

Statistical forecasting of the Australian macadamia crop

Three types of forecasts of the total Australian production of macadamia nuts (t nut-in-shell) have been produced early each year since 2001. The first is a long-term forecast, based on the expected production from the tree census data held by the Australian Macadamia Society, suitably scaled up for missing data and assumed new plantings each year. These long-term forecasts range out to 10 years in the future, and form a basis for industry and market planning. Secondly, a statistical adjustment (termed the climate-adjusted forecast) is made annually for the coming crop. As the name suggests, climatic influences are the dominant factors in this adjustment process, however, other terms such as bienniality of bearing, prices and orchard aging are also incorporated. Thirdly, industry personnel are surveyed early each year, with their estimates integrated into a growers and pest-scouts forecast. Initially conducted on a 'whole-country' basis, these models are now constructed separately for the six main production regions of Australia, with these being combined for national totals. Ensembles or suites of step-forward regression models using biologically-relevant variables have been the major statistical method adopted, however, developing methodologies such as nearest-neighbour techniques, general additive models and random forests are continually being evaluated in parallel. The overall error rates average 14% for the climate forecasts, and 12% for the growers' forecasts. These compare with 7.8% for USDA almond forecasts (based on extensive early-crop sampling) and 6.8% for coconut forecasts in Sri Lanka. However, our somewhatdisappointing results were mainly due to a series of poor crops attributed to human reasons, which have now been factored into the models. Notably, the 2012 and 2013 forecasts averaged 7.8 and 4.9% errors, respectively. Future models should also show continuing improvement, as more data-years become available.

Object detection and recognition in complex scenes

Dissertação de Mestrado, Engenharia Informática, Faculdade de Ciências e Tecnologia, Universidade do Algarve, 2014

Avaliação e controlo da qualidade da água de albufeiras com base em modelos dinâmicos

A problemática relacionada com a modelação da qualidade da água de albufeiras pode ser abordada de diversos pontos de vista. Neste trabalho recorre-se a metodologias de resolução de problemas que emanam da Área Cientifica da Inteligência Artificial, assim como a ferramentas utilizadas na procura de soluções como as Árvores de Decisão, as Redes Neuronais Artificiais e a Aproximação de Vizinhanças. Actualmente os métodos de avaliação da qualidade da água são muito restritivos já que não permitem aferir a qualidade da água em tempo real. O desenvolvimento de modelos de previsão baseados em técnicas de Descoberta de Conhecimento em Bases de Dados, mostrou ser uma alternativa tendo em vista um comportamento pró-activo que pode contribuir decisivamente para diagnosticar, preservar e requalificar as albufeiras. No decurso do trabalho, foi utilizada a aprendizagem não-supervisionada tendo em vista estudar a dinâmica das albufeiras sendo descritos dois comportamentos distintos, relacionados com a época do ano. ABSTRACT: The problems related to the modelling of water quality in reservoirs can be approached from different viewpoints. This work resorts to methods of resolving problems emanating from the Scientific Area of Artificial lntelligence as well as to tools used in the search for solutions such as Decision Trees, Artificial Neural Networks and Nearest-Neighbour Method. Currently, the methods for assessing water quality are very restrictive because they do not indicate the water quality in real time. The development of forecasting models, based on techniques of Knowledge Discovery in Databases, shows to be an alternative in view of a pro-active behavior that may contribute to diagnose, maintain and requalify the water bodies. ln this work. unsupervised learning was used to study the dynamics of reservoirs, being described two distinct behaviors, related to the time of year.

Our neighbour's view : Australian media cover of PNG

This chapter reviews the incidence of coverage of Papua New Guinea affairs in the Australian press and in Australian broadcast media. It presents the findings of a formal monitoring of selected newspaper coverage and news broadcasts of the leading Australian television and radio outlets. The study also includes news stories published on ABC Online. The findings for print media suggest that coverage of PNG is inadequate and may be contributing towards negative images of that country in Australia. The broadcast monitoring found also that beyond the ABC's regular and balanced coverage, there was very little mention of PNG on Australian airwaves. The deployment of resources by the ABC was seen as a potential model for increased quantity and quality of coverage, with its maintenance of a correspondent and office in the country, and use of reports from PNG across a wide range of programs. The investigation noted some early indications of a shift in media attention, following the election of a new government in Australia in 2007, which gave some priority attention to PNG including a visit by the then Prime Minister.

A common neighbour based two-way collaborative recommendation method

Traditional recommendation methods offer items, that are inanimate and one way recommendation, to users. Emerging new applications such as online dating or job recruitments require reciprocal people-to-people recommendations that are animate and two-way recommendations. In this paper, we propose a reciprocal collaborative method based on the concepts of users' similarities and common neighbors. The dataset employed for the experiment is gathered from a real life online dating network. The proposed method is compared with baseline methods that use traditional collaborative algorithms. Results show the proposed method can achieve noticeably better performance than the baseline methods.

Short-term traffic volume forecasting : a k-nearest neighbor approach enhanced by constrained linearly sewing principle component algorithm

To enhance the performance of the k-nearest neighbors approach in forecasting short-term traffic volume, this paper proposed and tested a two-step approach with the ability of forecasting multiple steps. In selecting k-nearest neighbors, a time constraint window is introduced, and then local minima of the distances between the state vectors are ranked to avoid overlappings among candidates. Moreover, to control extreme values’ undesirable impact, a novel algorithm with attractive analytical features is developed based on the principle component. The enhanced KNN method has been evaluated using the field data, and our comparison analysis shows that it outperformed the competing algorithms in most cases.

Improved image set classification via joint sparse approximated nearest subspaces

Existing multi-model approaches for image set classification extract local models by clustering each image set individually only once, with fixed clusters used for matching with other image sets. However, this may result in the two closest clusters to represent different characteristics of an object, due to different undesirable environmental conditions (such as variations in illumination and pose). To address this problem, we propose to constrain the clustering of each query image set by forcing the clusters to have resemblance to the clusters in the gallery image sets. We first define a Frobenius norm distance between subspaces over Grassmann manifolds based on reconstruction error. We then extract local linear subspaces from a gallery image set via sparse representation. For each local linear subspace, we adaptively construct the corresponding closest subspace from the samples of a probe image set by joint sparse representation. We show that by minimising the sparse representation reconstruction error, we approach the nearest point on a Grassmann manifold. Experiments on Honda, ETH-80 and Cambridge-Gesture datasets show that the proposed method consistently outperforms several other recent techniques, such as Affine Hull based Image Set Distance (AHISD), Sparse Approximated Nearest Points (SANP) and Manifold Discriminant Analysis (MDA).