Econometric and predictive modeling techniques are two popular forecasting techniques. Both ofthese techniques have their own advantages and disadvantages. In this thesis some econometricmodels are considered and compared to predictive models using sales data for five products fromICA a Swedish retail wholesaler. The econometric models considered are regression model,exponential smoothing, and ARIMA model. The predictive models considered are artificialneural network (ANN) and ensemble of neural networks. Evaluation metrics used for thecomparison are: MAPE, WMAPE, MAE, RMSE, and linear correlation. The result of this thesisshows that artificial neural network is more accurate in forecasting sales of product. But it doesnot differ too much from linear regression in terms of accuracy. Therefore the linear regressionmodel which has the advantage of being comprehensible can be used as an alternative to artificialneural network. The results also show that the use of several metrics contribute in evaluatingmodels for forecasting sales. / Program: Magisterutbildning i informatik

Third phase formation is an undesirable phenomenon during the PUREX process, which is a continuous liquid-liquid extraction approach for the reprocessing of uranium and plutonium from spent nuclear fuel. When third phase formation occurs, the organic extraction solution splits into two layers. The light upper layer, which is commonly named the light organic phase, contains a lower concentration of metal ions, tri-n-butyl phosphate (TBP) and nitric acids but is rich in the organic diluent. The heavy lower layer, which is commonly named the third phase, contains high concentrations of metal ions, TBP and nitric acids. As the third phase contains high concentrations of the uranium and plutonium complexes it can thus cause processing and safety concerns. Therefore, a comprehensive understanding of the mechanism of third phase formation is needed so as to improve the PUREX flowsheet. To investigate third phase formation through molecular simulations, one should first obtain reliable molecular models. A refined model for TBP, which uses a new set of partial charges generated from our density functional theory calculations, was proposed in this study. To compare its performance with other available TBP models, molecular dynamics simulations were conducted to calculate the thermodynamic properties, transport properties and the microscopic structures of liquid TBP, TBP/water mixtures and TBP/n-alkane mixtures. To our knowledge, it is only TBP model that has been validated to show a good prediction of the microscopic structure of systems that consist of both hydrophobic and hydrophilic species. This thesis also presents evidence that the light-organic/third phase transition in the TBP/n-dodecane/HNO3/H2O systems, which is relevant to the PUREX process, is an unusual transition between two isotropic, bi-continuous micro-emulsion phases. The light-organic /third phase coexistence was first observed using Gibbs Ensemble Monte Carlo (GEMC) simulations and then validated through Gibbs free energy calculations. Snapshots from the simulations as well as the cluster analysis of the light organic and third phases reveal structures akin to bi-continuous micro-emulsion phases, where the polar species reside within a mesh whose surface consists of amphiphilic TBP molecules. The non-polar n-dodecane molecules are outside this mesh. The large-scale structural differences between the two phases lie solely in the dimensions of the mesh. To our knowledge, the observation of the light-organic/third phase coexistence through simulation approaches and a phase transition of this nature have not previously been reported. Finally, this thesis presents evidence that the microscopic structure of the light organic phase of the Zr(IV)/TBP/n-octane/HNO3/H2O system, which is also related to the PUREX process, is different from that of the common hypothesis, where such system is consisted of large ellipsoidal reverse micelles. Snapshots from simulations, hydrogen bonding analysis and cluster analysis showed that the Zr4+, nitrate, TBP and H2O form extended aggregated networks. Thus, as above, we observe a bi-continuous structure but this time with embedded local clusters centred around the Zr4+ ions. The local clusters were found to consist primarily of Zr(NO3)4·3TBP complexes. This finding provides a new view of the structure of the Zr(IV)/TBP/n-octane/HNO3/H2O system.

A representação do Sistema de Regulação Gênica por meio de uma Rede de Regulação Gênica (GRN) pode facilitar a compreensão dos processos biológicos no nível molecular, auxiliando no entendimento do comportamento dos genes, a descoberta da causa de doenças e o desenvolvimento de novas drogas. Através das GRNs pode-se avaliar quais genes estão ativos e quais são suas influências no sistema. Nos últimos anos, vários métodos computacionais foram desenvolvidos para realizar a inferência de redes a partir de dados de expressão gênica. Esta pesquisa apresenta uma análise comparativa de métodos de inferência de GRNs, realizando uma revisão do modelo experimental descrito na literatura atual aplicados a conjuntos de dados contendo poucas amostras. Apresenta também o uso comitês de especialistas (ensemble) para agregar o resultado dos métodos a fim de melhorar a qualidade da inferência. Como resultado obteve-se que o uso de poucas amostras de dados (abaixo de 50) não fornecem resultados interessantes para a inferência de redes. Demonstrou-se também que o uso de comitês de especialistas melhoram os resultados de inferência. Os resultados desta pesquisa podem auxiliar em pesquisas futuras baseadas em GRNs. / The representation of the gene regulation system by means of a Gene Regulatory Network (GRN) can help the understanding of biological processes at the molecular level, elucidating the behavior of genes and leading to the discovery of disease causes and the development of new drugs. GRNs allow to evaluate which genes are active and how they influence the system. In recent years, many computational methods have been developed for networks inference from gene expression data. This study presents a comparative analysis of GRN inference methods, reviewing the experimental modeling present in the state-of-art scientific publications applied to datasets with small data samples. The use of ensembles was proposed to improve the quality of the network inference. As results, we show that the use of small data samples (less than 50 samples) do not show a good result in the network inference problem. We also show that the use of ensemble improve the network inference.

Combining embedded systems and machine learning models is an exciting prospect. However, to fully target any embedded system, with the most stringent resource requirements, the models have to be designed with care not to overwhelm it. Decision tree ensembles are targeted in this thesis. A benchmark model is created with LightGBM, a popular framework for gradient boosted decision trees. This model is first transformed and regularized with RuleFit, a LASSO regression framework. Then it is further optimized with quantization and weight sharing, techniques used when compressing neural networks. The entire process is combined into a novel framework, called ESRule. The data used comes from the domain of frequency measurements in cellular networks. There is a clear use-case where embedded systems can use the produced resource optimized models. Compared with LightGBM, ESRule uses 72ˆ less internal memory on average, simultaneously increasing predictive performance. The models use 4 kilobytes on average. The serialized variant of ESRule uses 104ˆ less hard disk space than LightGBM. ESRule is also clearly faster at predicting a single sample.

Cette étude a pour objectif d’éclairer la notion d’ensemble de nouvelles au sein de l’œuvre de Marcel Arland. Couronné par le prix Goncourt pour son roman L’Ordre, cet auteur a abandonné ce genre pour choisir la nouvelle puis a revendiqué le recueil ensemble. En nous appuyant sur un corpus qu’il a lui même considéré comme un triptyque composant un vitrail : Les plus beaux de nos Jours, Il faut de tout pour faire un Monde et Les Vivants, nouscroiserons approches narratologique, onomastique et réceptive pour mettre à jour les récurrences qui se tissent de nouvelles en nouvelles.Celles cisont discrètes et ne s’appuient pas sur le retour d’un personnage principal identifiable. Entremêlant traumas biographiques et récit fictionnel, Arland invite le lecteur à découvrir les instants de grâce que connaissent d’humbles personnages aux existences souvent morbides. Sous une plume au classicisme impeccable, Arland instille ironie et tragédie au sein de ses nouvelles pour laisser entrevoir au lecteur attentif un monde que le temps apaise enfin. Le recueil ensemble permet à Arland de construire un monde au sein duquel morts et vivants s’unissent en un chant commun. / The aim of this study is to enlighten the topic of «  Whole  of  short  stories  »  within  the  Marcel Arland’s works. Awarded with the « Prix Goncourt » thanks to his novel L’Ordre,  this author abandoned this literary genre, chose instead of it the «  short­story » one and finally claimed the « collection­whole ». Using  a corpus which he himself considered as a triptych composing a  stained glass­window : ʺLes plus beaux de nos Joursʺ, ʺIl faut de tout  pour  faire  un  Mondeʺ  et  ʺLes  Vivantsʺ,  we  will  intertwine  the  narratological,  onomastic  and  receptive  approaches  in  order  to  uncover the recurrences which are forged from short-story to shortstory. These are unobtrusive and they don’t use the return of an  identifiable  main  character.  Intermingling  biographical  traumas  and fictional narrative, Arland invites his reader to discover the moments of grace that humble characters, with, often, morbid existences, are living. Under a faultless classical quill, Arland instils  irony  and  tragedy  within  his  short­stories  in  order to let the attentive reader glimpse a world that time at last soothes. The «  collection­whole » allows Arland to build a world in which the living and the dead become united in a common singing.

[Truncated abstract] The present study investigated whether any of the characteristics of the compound action potential (CAP) waveform or the spectrum of the neural noise (SNN) recorded from the cochlea, could be used to examine abnormal spike generation in the type I primary afferent neurones, possibly due to pathologies leading to abnormal hearing such as tinnitus or tone decay. It was initially hypothesised that the CAP waveform and SNN contained components produced by the local action currents generated at the peripheral ends of the type I primary afferent neurones, and that changes in these local action currents occurred due to changes in the membrane potential of these neurones. It was further hypothesised that the lateral olivo-cochlear system (LOCS) efferent neurones regulate the membrane potential of the primary afferent dendrites to maintain normal action potential generation, where instability in the membrane potential might lead to abnormal primary afferent firing, and possibly one form of tinnitus. We had hoped that the activity of the LOCS efferent neurones could be observed through secondary changes in the CAP waveform and SNN, resulting from changes in the membrane potential of the primary afferent neurones. The origins of the neural activity generating the CAP waveform and SNN peaks, and the effects of the LOCS on the CAP and SNN were experimentally investigated in guinea pigs using lesions in the auditory system, transient ischemia and asphyxia, focal and systemic temperature changes, and pharmacological manipulations of different regions along the auditory pathway. ... Therefore, the CAP and SNN are altered by changes in the propagation of the action potential along the primary afferent neurones, by changes in the morphology of the tissues surrounding the cochlear nerve, and by changes in the time course of the action currents. If the CAP waveform is not altered, the amplitude of the 1kHz speak in the spontaneous SNN can be used as an objective measure of the spontaneous firing rate of the cochlear neurones. However, because the SNN contains a complex mixture of neural activity from all cochlear neurones, and the amplitude of the spontaneous SNN is variable, it would be difficult to use the spontaneous SNN alone as a differential diagnostic test of cochlear nerve pathologies. To record extratympanic electrocochleography (ET ECochG) from humans, a custom-designed, inexpensive, low-noise, optically isolated biological amplifier was built. Furthermore, a custom-designed extratympanic active electrode and ear canal indifferent electrode were designed, which increased the signal-to-noise ratio of the ECochG recording by a factor of 2, decreasing the overall recording time by 75%. The human and guinea pig CAP waveforms recorded in the present study appeared similar, suggesting that the origins of the human and guinea pig CAP waveforms were the same, and that experimental manipulations of the guinea pig CAP waveform can be used to diagnose the cause of abnormal human ECochG waveforms in cases of cochlear nerve pathologies.

This thesis advances the understanding of the application of artificial neural networks ensemble to clinical data by addressing the following fundamental question: What is the potentiality of an ensemble of neural networks models as a filter and classifier in a complex clinical situation? A novel neural networks ensemble classification model called Rules and Information Driven by Consistency in Artificial Neural Networks Ensemble (RIDCANNE) is developed for the purpose of prediction of medical outcomes or events, such as kidney transplants. The proposed classification model is based on combination of initial data preparations, preliminary classification by ensembles of Neural Networks, and generation of new training data based on criteria of highly accuracy and model agreement. Furthermore, it can also generate decision tree classification models to provide classification of data and the prediction results. The case studies described in this thesis are from a kidney transplant database and two well-known collections of benchmark data known as the Pima Indian Diabetes and Wisconsin Cancer datasets. An implication of this study is that further attention needs to be given to both data collection and preparation stages. This study revealed that even neural network ensemble models that are known for their strong generalization ability might not be able to provide a high level of accuracy for complex, noisy and incomplete clinical data. However, by using a selective subset of data points, it is possible to improve the overall accuracy. In summary, the research conducted for this thesis advances the current clinical data preparation and classification techniques in which the task is to extract patterns that contain higher information content from a sea of noisy and incomplete clinical data, and build accurate and transparent classifiers. The RIDC-ANNE approach improves an analyst�s ability to better understand the data. Furthermore, it shows great promise for use in clinical decision making systems. It can provide us with a valuable data mining tool with great research and commercial potential.

Thesis (M. Phil.)--University of Hong Kong, 2007. / Title proper from title frame. Also available in printed format.

La comparaison de séquences est une des tâches fondamentales de la bioinformatique. Les nouvelles technologies de séquençage conduisent à une production accélérée des données génomiques et renforcent les besoins en outils rapides et efficaces pour effectuer cette tâche. Dans cette thèse, nous proposons un nouvel algorithme de comparaison intensive de séquences, explicitement conçu pour exploiter toutes les formes de parallélisme présentes dans les microprocesseurs de dernière génération (instruction SIMD, architecture multi-coeurs). Cet algorithme s'adapte également à un parallélisme massif que l'on peut trouver sur des accélérateurs de type FPGA ou GPU. Cet algorithme a été mis en oeuvre à travers le logiciel PLAST (Parallel Local Alignment Search Tool). Différentes versions sont disponibles suivant les données à traiter (protéine et/ou ADN). Une version MPI a également été mise au point pour un déploiement sur un cluster de PCs. En fonction de la nature des données et des technologies employées des accélérations de 3 à 20 ont été mesurées par rapport à la référence du domaine, le logiciel BLAST, pour un niveau de qualité équivalent.

Divers algorithmes d'études locale d'ensembles invariants compacts de systèmes dynamiques sont présentés dans ce travail. Nous commençons par développer des méthodes de calcul numérique de la densité locale autour d'un point d'un ensemble de Julia de fraction rationnelle. Ce problème est important dans le domaine de l'étude des modèles hiérarchiques de la physique statistique. De plus, cette densité serait un des paramètres principaux de la caractérisation d'invariants compacts de systèmes dynamiques (attracteur étrange...). L'application de ces méthodes demande des algorithmes d'accès rapide à des régions (rectangle, cercle) du compact numériquement approche par un ensemble forme d'un grand nombre de points. Nous avons mis au point un algorithme, réellement implémentable et expérimentalement efficace, qui résout ce problème. Nous montrons que, sous certaines conditions, ce procédé permet l'estimation de quelques dimensions fractales de l'ensemble considère. Un logiciel, nomme Elsep et écrit en langage Pascal, qui regroupe et exploite tous ces algorithmes ponctue cette étude. Des résultats numériques et graphiques illustrent chacune des parties traitées