Automatical electrochemical measurements are sources of large data sets intended for further analysis. This work deals with classification, evaluation and processing of electrochemical signals using artificial neural networks. Due to high dimensionality of input data, an autoassociative neural network (AANN) is used in this work. This type of network performs dimensionality reduction via filtering the input data into relatively small number of principal parameters at the bottleneck output. These extracted parameters can be used for classification, evaluation and additional modelling of analyzed data trough the reconstructive part of this network. Furthermore, this work deals with implementation of a feedforward neural network in OpenCL language.

This paper deals with the classification of ECG measured from isolated rabbit hearts during the experiment with repeated ischemia. Classification features were calculated using the methods of heart rate variability analysis. The results were statistically evaluated. Heart rate variability parameters were calculated using Kubios HRV, other calculations were performed in MATLAB. Artificial neural network was created to classify the analyzed parameters to specific groups.

The goal of this thesis is to study the possibilities of Artificial neural network as an innovative mathematical methods for financial analysis of company performance, to find out what are today´s requests for performance evaluation of companies are and to identify possible ways how to use this relatively new concept in this area. When processing the possibilities of the computer program Maple for mathematical calculations will be applied. Intermediate objectives are then acquainted with the basic principle on which the artificial neural networks works, to analyze the financial performance of specific company and evaluate potential predictive abilities of the proposed network. The result of the work should be evaluating the success of this approach to financial analysis and evaluation of its use in practice.

In the evaluation of special populations, such as neonates, infants and malingerers, audiologist often have to rely heavily on objective measurements to assess hearing ability. Current objective audiological procedures such as tympanometry, the acoustic reflex, auditory brainstem response and transient evoked otoacoustic emissions, however, have certain limitations, contributing to the need of an objective, non¬invasive, rapid, economic test of hearing that evaluate hearing ability in a wide range of frequencies. The purpose of this study was to investigate distortion product otoacoustic emissions (DPOAEs) as an objective test of hearing. The main aim was to attempt to predict hearing ability at 500 Hz, 1000 Hz, 2000 Hz and 4000 Hz with DPOAEs and artificial neural networks (ANNs) in normal and hearing-impaired ears. Other studies that attempted to predict hearing ability with DPOAEs and conventional statistical methods were only able to distinguish between normal and impaired hearing. Back propagation neural networks were trained with the pattern of all present and absent DPOAE responses of 11 DPOAE frequencies of eight DP Grams and pure tone thresholds at 500 Hz, 1000 Hz, 2000 Hz and 4000 Hz. The neural network used the learned correlation between these two data sets to predict hearing ability at 500 Hz, 1000 Hz, 2000 Hz and 4000 Hz. Hearing ability was not predicted as a decibel value, but into one of several categories spanning 10-15dB. Results indicated that prediction accuracy of normal hearing was 92% at 500 Hz, 87% at 1000 Hz, 84% at 2000 Hz and 91% at 4000 Hz. The prediction of hearing-impaired categories was less satisfactory, due to insufficient data for the ANNs to train on. The variables age and gender were included in some of the neural network runs to determine their effect on the distortion product. Gender had only a minor positive effect on prediction accuracy, but age affected prediction accuracy considerably in a positive way. The effect of the amount of data that the neural network had to train on was also investigated. A prediction versus ear count correlation strongly suggested that the inaccurate predictions of hearing-impaired categories is not a result of an inability of DPOAEs to predict pure tone thresholds in hearing impaired ears, but a result of insufficient data for the neural network to train on. This research concluded that DPOAEs and ANNs can be used to accurately predict hearing ability within 10dB in normal and hearing-impaired ears from 500 Hz to 4000 Hz for hearing losses of up to 65dB HL. / Dissertation (MCommunication Pathology)--University of Pretoria, 2009. / Speech-Language Pathology and Audiology / unrestricted

This thesis conducts a study on a data set from the Swedish and Danish postal service Postnord, comparing an artificial neural network (ANN) and a gradient boosting machine (GBM) for predicting delays in package deliveries. The models are evaluated based on F1-score for the important class which represents the data points that are delayed and needed to be identified. The GBM is already implemented and tuned using grid search by Postnord, the ANN is tuned using sequential model based optimization with the tree Parzen estimator function. Furthermore, it is trained using dynamic resampling to handle the imbalanced data set. Even with several measures implemented to handle the class imbalance, the ANN performs poorly when tested on unseen data, unlike the GBM. The GBM has high precision (84%) and decent recall (24%), which produces a F1-score of 0.38. The ANN has high recall (62%) but extremely low precision (5%) which gives a F1-score of 0.08, indicating that it is biased to predict sample as delayed when it is in time. The GBM has a natural handling of class imbalance unlike the ANN, and even with measures taken to improve the ANN and its handling of class imbalance, GBM performs better.

Autoscalers handle the scaling of instances in a system automatically based on specified thresholds such as CPU utilization. Reactive autoscalers do not take the delay of initiating a new instance into account, which may lead to overutilization. By applying machine learning methodology to predict future loads and the desired number of instances, it is possible to preemptively initiate scaling such that new instances are available before demand occurs. Leveraging efficient scaling policies keeps the costs and energy consumption low while ensuring the availability of the system. In this thesis, the predictive capability of different multilayer perceptron configurations is investigated to elicit a suitable model for a telecom support system. The results indicate that it is possible to accurately predict future load using a multilayer perceptron regressor model. However, the possibility of reproducing the results in a live environment is questioned as the dataset used is derived from a simulation.

L’objectif de la thèse est d’étudier la gestion de bout en bout des architectures à la SDN, et comment nos briques OSS (Operation Support System) doivent évoluer: cela implique d’étudier les processus métier associés, leurs implémentations ainsi que l’outillage nécessaire. Les objectifs de la thèse sont donc de répondre aux verrous suivants:1. Identifier les changements impliqués par l’émergence de ces réseaux programmables sur les architectures de gestions en termes d’exigences ou "requirements". L’étude peut être focalisée sur un type de réseau, mobile par exemple. 2. Identifier l’évolution à apporter aux interfaces de gestions actuelles: quelles alternatives aux FCAPS (fault, configuration, accounting, performance, and security) ? Quels changements à apporter aux couches de gestions allant du gestionnaire d’équipement ou "Element Management System" jusqu’au OSS ? / The main goal of the PhD activities is to define and develop architecture and mechanisms to ensure consistency and continuity of the operations and behaviors in mixed physical/virtual environments, characterized by a high level of dynamicity, elasticity and heterogeneity by applying a cognitive approach to the architecture where applicable. The target is then to avoid the "build it first, manage it later" paradigm. The research questions targeted by the PhD are the following: 1. Identify the changes on Network Operation Support Systems implementation when using SDN as a design approach for future networks. The study could be restricted to mobile networks for example, or sub-part of it (CORE networks, RAN, data centers, etc); 2.Identify the needed evolution at the management interfaces level: a. Shall we need alternative to the well-known FCAPS and do we still need the element management system? b. What will change to provision an SDN based service? c. How to ensure resiliency of SDN based networks?

M. Tech. (Department of Information Technology, Faculty of Applied and Computer Sciences) Vaal University of Technology. / Cloud computing provides improved and simplified IT management and maintenance capabilities through central administration of resources. Companies of all shapes and sizes are adapting to this new technology. Although cloud computing is an attractive concept to the business community, it still has some challenges such as traffic management and traffic prediction that need to be addressed. Most cloud service providers experience traffic congestion. In the absence of effective tools for cloud computing traffic prediction, the allocation of resources to clients will be ineffective thus driving away cloud computing users. This research intends to mitigate the effect of traffic congestion on provision of cloud service by proposing a proactive traffic prediction model that would play an effective role in congestion control and estimation of accurate future resource demand. This will enhance the accuracy of traffic flow prediction in cloud computing by service providers. This research will evaluate to determine the performance between Auto-regressive Integrated Moving Average (ARIMA) and Artificial Neural Networks (ANN) as prediction tools for cloud computing traffic. These two techniques were tested by using simulation to predict traffic flow per month and per year. The dataset was downloaded data taken from CAIDA database. The two algorithms Auto-Regressive Integrated Moving Average (ARIMA) and Artificial Neural Networks (ANN) where implemented and tested separately. Experimental results were generated and analyzed to test the effectiveness of the traffic prediction algorithms. Finally, the findings indicated that ARIMA can have 98 % accurate prediction results while ANN produced 89 % accurate prediction results. It was also observed that both models perform better on monthly data as compared to yearly data. This study recommends ARIMA algorithm for data flow prediction in private cloud computing

Targeting heavier freights and transporting passengers with higher speeds became the strategic railway development during the past decades significantly increasing interests on railway networks. Among different components of a railway network, bridges constitute a major portion imposing considerable construction and maintenance costs. On the other hand, heavier axle loads and higher trains speeds may cause resonance occurrence on bridges; which consequently limits operational train speed and lines. Therefore, satisfaction of new expectations requires conducting a large number of dynamic assessments/analyses on bridges, especially on existing ones. Evidently, such assessments need detailed information, expert engineers and consuming considerable computational costs. In order to save the computational efforts and decreasing required amount of expertise in preliminary evaluation of dynamic responses, predictive models using artificial neural network (ANN) are proposed in this study. In this regard, a previously developed closed-form solution method (based on solving a series of moving force) was adopted to calculate the dynamic responses (maximum deck deflection and maximum vertical deck acceleration) of randomly generated bridges. Basic variables in generation of random bridges were extracted both from literature and geometrical properties of existing bridges in Sweden. Different ANN architectures including number of inputs and neurons were considered to train the most accurate and computationally cost-effective mode. Then, the most efficient model was selected by comparing their performance using absolute error (ERR), Root Mean Square Error (RMSE) and coefficient of determination (R2). The obtained results revealed that the ANN model can acceptably predict the dynamic responses. The proposed model presents Err of about 11.1% and 9.9% for prediction of maximum acceleration and maximum deflection, respectively. Furthermore, its R2 for maximum acceleration and maximum deflection predictions equal to 0.982 and 0.998, respectively. And its RMSE is 0.309 and 1.51E-04 for predicting the maximum acceleration and maximum deflection prediction, respectively. Finally, sensitivity analyses were conducted to evaluate the importance of each input variable on the outcomes. It was noted that the span length of the bridge and speed of the train are the most influential parameters.

Machine learning models are becoming popular in the industry since the technology has developed to solve numerous problems, such as classification [1], detection [2], and segmentation [3]. These algorithms require training with a large dataset which includes correct class labels to perform well on unseen data. One way to get access to large sets of annotated data is to use data from simulation engines. However this data is often not as complex and rich as real data, and for images, for examples, there can be a need to make these look more photorealistic. One approach to do this is denoted Domain adaptation. In collaboration with SAAB Aeronautics, which funds this research, this study aims to explore available domain adaptation frameworks, implement a framework and use it to make a transformation from simulation to real- life. A state-of-the-art framework CyCADA was re-implemented from scratch using Python and TensorFlow as a Deep Learning package. The CyCADA implementation was successfully verified by reproducing the digit adaptation result demonstrated in the original paper, making domain adaptations between MNIST, USPS, and SVHN. CyCADA was used to domain adapt landscape images from simulation to real-life. Domain-adapted images were used to train an object detector to evaluate whether CyCADA allows a detector to perform more accurately in real-life data. Statistical measurements, unfortunately, showed that domain-adapted images became less photorealistic with CyCADA, 88.68 FID on domain-adapted images compared to 80.43 FID on simulations, and object detection performed better on real-life data without CyCADA, 0.131 mAP with a detector trained on domain-adapted images compared to 0.681 mAP with simulations. Since CyCADA produced effective domain adaptation results between digits, there remains a possibility to try multiple hyperparameter settings and neural network architecture to produce effective results with landscape images. / Denna studie genomfördes i ett samarbete med SAAB Aeronautics och handlar om att utveckla en Domain Adaptation-modul som förbättrar prestandan av ett nätverk för objektdetektering. När ett objektdetekteringsnätverk är tränat med data från en domän så är det inte givet att samma nätverk presterar bra på en annan domän. Till exempel, ritningar och fotografier av frukter. Forskare löser problemet genom att samla data från varje domän och träna flera maskininlärningsalgoritmer, vilket är en lösning som kräver tid och energi. Detta problem kallas för domänskiftesproblem. Ett hett ämne inom djupinlärning handlar om att lösa just detta problem med domänskift och det finns en rad algoritmer som faller i kategorin Domain Adaptation. Denna studie utvecklar CyCADA som metod att evaluera en toppmodern Domain Adaptation-algoritm. Återimplementering av CyCADA blev lyckad, eftersom flera resultat var återskapade från den originala artikeln. CyCADA producerade effektiva domänskiften på bilder av siffror. CyCADA användes med landskapsbilder från en simulator för att öka verklighetsförankringen på bilderna. Domänskiftade landskapsbilder blev suddiga med CyCADA. FID värdet av domänskiftade bilder, ett utvärderingsmått som evaluerar fotorealism av bilder, blev lägre i jämförelse med endast simulerade bilder. Objektdetekteringsnätverket presterade bättre utan användning av CyCADA. Givet att CyCADA presterade bra i att transformera bilder av siffror från en domän till en annan finns det hopp om att ramverket kan prestera bra med landskapsbilder med fler försök i att ställa in hyperparameterar.