<p dir="ltr">Impaired cerebral vasoregulation is one of the most common post-ischemic stroke effects. Diagnosis and prevention of this condition is often invasive, costly and in-effective. This impairment restricts the cerebral blood vessels to properly regulate blood flow, which is very important for normal brain functioning. Developing accurate, non-invasive and efficient methods to detect this condition aids in better stroke diagnosis and prevention. </p><p dir="ltr">The aim of this thesis is to develop deep learning techniques for the purpose of detection of cerebral vasoregulation impairments by analyzing physiological signals. This research employs various Deep learning techniques like Convolution Neural Networks (CNN), MobileNet, and Long-Short-Term Memory (LSTM) to determine variety of physiological signals from the PhysioNet database like Electrocardio-gram (ECG), Transcranial Doppler (TCD), Electromyogram (EMG), and Blood Pressure(BP) as stroke or non-stroke subjects. The effectiveness of these algorithms is demonstrated by a classification accuracy of 90\% for the combination of ECG and EMG signals. </p><p dir="ltr">Furthermore, this research explores the importance of analyzing dynamic physiological activities in determining the impairment. The dynamic activities include Sit-stand, Sit-stand-balance, Head-up-tilt, and Walk dataset from the PhysioNet website. CNN and MobileNetV3 are employed in classification purposes of these signals, attempting to identify cerebral health. The accuracy of the model and robustness of these methods is greatly enhanced when multiple signals are integrated. </p><p dir="ltr">Overall, this study highlights the potential of deep multimodal physiological learning in the development of precision brain medicine further enhancing stroke diagnosis. The results pave the way for the development of advanced diagnostic tools to determine cerebral health. </p>

<p dir="ltr">This study aims to develop Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU), and Transformer-based Informer models and evaluate the performance of these models using data from one, two, three, and four weeks in advance to predict the progression of rice leaf blast disease; and assess the generalizability of these models across various climatic regions in Taiwan. This research utilized multi-location rice leaf blast diseased leaf percentage data collected between 2015 and 2021 in Taiwan, along with weather data from the Taiwanese meteorological observation network to predict rice blast disease one week in advance, serving as a benchmark for comparing with predictions made two, three, and four weeks in advance.</p>

京都大学 / 新制・課程博士 / 博士(工学) / 甲第24895号 / 工博第5175号 / 新制||工||1988(附属図書館) / 京都大学大学院工学研究科社会基盤工学専攻 / (主査)教授 KIM Chul-Woo, 教授 杉浦 邦征, 教授 八木 知己 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

<p>Our modern society is more dependent on its assets in space now more than ever. For<br> example, the Global Positioning System (GPS) many rely on for navigation uses data from a<br> 24-satellite constellation. Additionally, our current infrastructure for gas pumps, cell phones,<br> ATMs, traffic lights, weather data, etc. all depend on satellite data from various constel-<br> lations. As a result, it is increasingly necessary to accurately track and predict the space<br> domain. In this thesis, after discussing how space object tracking and object position pre-<br> diction is currently being done, I propose a machine learning-based approach to improving<br> the space object position prediction over the standard SGP4 method, which is limited in<br> prediction accuracy time to about 24 hours. Using this approach, we are able to show that<br> meaningful improvements over the standard SGP4 model can be achieved using a machine<br> learning model built based on a type of recurrent neural network called a long short term<br> memory model (LSTM). I also provide distance predictions for 4 different space objects over<br> time frames of 15 and 30 days. Future work in this area is likely to include extending and<br> validating this approach on additional satellites to construct a more general model, testing a<br> wider range of models to determine limits on accuracy across a broad range of time horizons,<br> and proposing similar methods less dependent on antiquated data formats like the TLE.</p>

Cellular network traffic prediction is a critical challenge for communication providers, which is important for use cases such as traffic steering and base station resources management. Traditional prediction methods mostly rely on historical time-series data to predict traffic load, which often fail to model the real world and capture surrounding environment conditions. In this work, we propose a multi-modal deep learning model for 4G/5G Cellular Network Traffic prediction by considering external data sources such as satellite imagery and weather data. Specifically, our proposed model consists of three components (1) temporal component (modeling correlations between traffic load values with historical data points via LSTM) (2) computer vision component (using embeddings to capture correlations between geographic regions that share similar landscape patterns using satellite imagery data and state of the art CNN models), and (3) weather component (modeling correlations between weather measurements and traffic patterns). Furthermore, we study the effects and limitations of using such contextual datasets on time series learning process. Our experiments show that such hybrid models do not always lead to better performance, and LSTM model is capable of modeling complex sequential interactions. However, there is a potential for classifying or labelling regions by their urban landscape and the network traffic. / Förutsägelse av mobilnätstrafik är en kritisk utmaning för kommunikation leverantörer, där användningsområden inkluderar trafikstyrning och hantering av basstationsresurser. Traditionella förutsägelsesmetoder förlitar sig främst på historisk tidsseriedata för att förutsäga trafikbelastning, detta misslyckas ofta med att modellera den verkliga världen och fånga omgivande miljö. Det här arbetet föreslår en multimodal modell med djupinlärning förutsägelse av 4G/5G nätverkstrafik genom att beakta externa datakällor som satellitbilder och väderdata. Specifikt består vår föreslagna modell av tre komponenter (1) temporal komponent (korrelationsmodellering mellan trafikbelastningsvärden med historiska datapunkter via LSTM) (2) datorseende komponent (med inbäddningar för att fånga korrelationer mellan geografiska regioner som delar liknande landskapsmönster med hjälp av satelitbilddata och state-of-the-art CNN modeller), och (3) väderkomponent (modellerande korrelationer mellan vädermätningar och trafikmönster). Dessutom studerar vi effekterna och begränsningarna av att använda sådana kontextuella datamängder på tidsserieinlärningsprocessen. Våra experiment visar att hybridmodeller inte alltid leder till bättre prestanda och att LSTM-modellen är kapabel att modellera komplexa sekventiella interaktioner. Det finns dock en potential att klassificera eller märka regioner efter deras stadslandskap och nättrafiken. / La prévision du trafic sur les réseaux cellulaires est un défi crucial pour les fournisseurs de communication, ce qui est important pour les cas d’utilisation tels que la direction du trafic et la gestion des ressources des stations de base. Les méthodes de prédiction traditionnelles reposent principalement sur des données historiques de séries chronologiques pour prédire la charge de trafic, qui échouent souvent à modéliser le monde réel et à capturer les conditions de l’environnement environnant. Dans ce travail, nous proposons un modèle d’apprentissage profond multimodal pour la prédiction du trafic des réseaux cellulaires 4G/5G en considérant des sources de données externes telles que l’imagerie satellitaire et les données météorologiques. Plus précisément, notre modèle proposé se compose de trois composants (1) composant temporel (modélisation des corrélations entre les valeurs de charge de trafic avec des points de données historiques via LSTM) (2) composant de vision par ordinateur (utilisant des incorporations pour capturer les corrélations entre les régions géographiques qui partagent des modèles de paysage similaires à l’aide de données d’imagerie satellitaire et de modèles CNN de pointe) et (3) composante météorologique (modélisation des corrélations entre les mesures météorologiques et les modèles de trafic). De plus, nous étudions les effets et les limites de l’utilisation de tels ensembles de données contextuelles sur le processus d’apprentissage des séries chronologiques. Nos expériences montrent que de tels modèles hybrides ne conduisent pas toujours à de meilleures performances, et le modèle LSTM est capable de modéliser des interactions séquentielles complexes. Cependant, il est possible de classer ou d’étiqueter les régions en fonction de leur paysage urbain et du trafic du réseau.

Current approaches to generating rhyming English poetry with a neural network involve constraining output to enforce the condition of rhyme. We investigate whether this approach is necessary, or if recurrent neural networks can learn rhyme patterns on their own. We compile a new dataset of amateur poetry which allows rhyme to be learned without external constraints because of the dataset’s size and high frequency of rhymes. We then evaluate models trained on the new dataset using a novel framework that automatically measures the system’s knowledge of poetic form and generalizability. We find that our trained model is able to generalize the pattern of rhyme, generate rhymes unseen in the training data, and also that the learned word embeddings for rhyming sets of words are linearly separable. Our model generates a couplet which rhymes 68.15% of the time; this is the first time that a recurrent neural network has been shown to generate rhyming poetry a high percentage of the time. Additionally, we show that crowd-source workers can only distinguish between our generated couplets and couplets from our dataset 63.3% of the time, indicating that our model generates poetry with coherency, semantic meaning, and fluency comparable to couplets written by humans. / Graduate

In this paper a novel speaker recognition system is introduced. Automated speaker recognition has become increasingly popular to aid in crime investigations and authorization processes with the advances in computer science. Here, a recurrent neural network approach is used to learn to identify ten speakers within a set of 21 audio books. Audio signals are processed via spectral analysis into Mel Frequency Cepstral Coefficients that serve as speaker specific features, which are input to the neural network. The Long Short-Term Memory algorithm is examined for the first time within this area, with interesting results. Experiments are made as to find the optimum network model for the problem. These show that the network learns to identify the speakers well, text-independently, when the recording situation is the same. However the system has problems to recognize speakers from different recordings, which is probably due to noise sensitivity of the speech processing algorithm in use.

Streaming media is a growing market all over the world which sets a strict requirement on mobile connectivity. The foundation for a good user experience when supplying a streaming media service on a mobile device is to ensure that the user can access the requested content. Due to the varying availability of mobile connectivity measures has to be taken to remove as much dependency as possible on the quality of the connection. This thesis investigates the use of a Long Short-Term Memory machine learning model for predicting a future geographical location for a mobile device. The predicted location in combination with information about cellular connectivity in the geographical area is used to schedule prefetching of media content in order to improve user experience and to reduce mobile data usage. The Long Short-Term Memory model suggested in this thesis achieves an accuracy of 85.15% averaged over 20000 routes and the predictive caching managed to retain user experience while decreasing the amount of data consumed. / <p>This thesis is written as a joint thesis between two students from different universities. This means the exact same thesis is published at two universities (LiU and KTH) but with different style templates. The other report has identification number: TRITA-EECS-EX-2018:403</p>

Submitted by Ronaldo Cesar Dametto (rdametto@uol.com.br) on 2018-09-18T19:17:34Z No. of bitstreams: 1 Dissertação_Completa_Final.pdf: 2885777 bytes, checksum: 05b2d5417efbec72f927cf8a62eef3fb (MD5) / Approved for entry into archive by Lucilene Cordeiro da Silva Messias null (lubiblio@bauru.unesp.br) on 2018-09-20T12:19:07Z (GMT) No. of bitstreams: 1 dametto_rc_me_bauru.pdf: 2877027 bytes, checksum: cee33d724090a01372e1292109af2ce9 (MD5) / Made available in DSpace on 2018-09-20T12:19:07Z (GMT). No. of bitstreams: 1 dametto_rc_me_bauru.pdf: 2877027 bytes, checksum: cee33d724090a01372e1292109af2ce9 (MD5) Previous issue date: 2018-08-06 / O aprendizado de máquina vem sendo utilizado em diferentes segmentos da área financeira, como na previsão de preços de ações, mercado de câmbio, índices de mercado e composição de carteira de investimento. Este trabalho busca comparar e combinar três tipos de algoritmos de aprendizagem de máquina, mais especificamente, o método Ensemble de Redes Neurais Artificias com as redes Multilayer Perceptrons (MLP), auto-regressiva com entradas exógenas (NARX) e Long Short-Term Memory (LSTM) para predição do Índice Bovespa. A amostra da série do Ibovespa foi obtida pelo Yahoo!Finance no período de 04 de janeiro de 2010 a 28 de dezembro de 2017, de periodicidade diária. Foram utilizadas as séries temporais referentes a cotação do Dólar, além de indicadores numéricos da Análise Técnica como variáveis independentes para compor a predição. Os algoritmos foram desenvolvidos através da linguagem Python usando framework Keras. Para avaliação dos algoritmos foram utilizadas as métricas de desempenho MSE, RMSE e MAPE, além da comparação entre as previsões obtidas e os valores reais. Os resultados das métricas indicam bom desempenho de predição pelo modelo Ensemble proposto, obtendo 70% de acerto no movimento do índice, porém, não conseguiu atingir melhores resultados que as redes MLP e NARX, ambas com 80% de acerto. / Different segments of the financial area, such as the forecast of stock prices, the foreign exchange market, the market indices and the composition of investment portfolio, use machine learning. This work aims to compare and combine two types of machine learning algorithms, the Artificial Neural Network Ensemble method with Multilayer Perceptrons (MLP), auto-regressive with exogenous inputs (NARX) and Long Short-Term Memory (LSTM) for prediction of the Bovespa Index. The Bovespa time series samples were obtained daily, using Yahoo! Finance, from January 4th, 2010 to December 28th, 2017. Dollar quotation, Google trends and numerical indicators of the Technical Analysis were used as independent variables to compose the prediction. The algorithms were developed using Python and Keras framework. Finally, in order to evaluate the algorithms, the MSE, RMSE and MAPE performance metrics, as well as the comparison between the obtained predictions and the actual values, were used. The results of the metrics indicate good prediction performance by the proposed Ensemble model, obtaining a 70% accuracy in the index movement, but failed to achieve better results than the MLP and NARX networks, both with 80% accuracy.

Cryptocurrencies are a cryptography based technology, that has increased massively in popularity in recent years. These currencies are traded on markets that specialize in cryptocurrency trade. There, you can trade one cryptocurrency for another, or buy one with real world money. These markets are quite volatile, meaning that the price of most cryptocurrencies swing up and down a lot. The largest cryptocurrency is Bitcoin, but there is also more than 1500 smaller ones, that goes by the name alternative coins, or altcoins. This thesis will try to find out if it is possible to make accurate predictions about the future price of the altcoin Ethereum, and also see if Bitcoin may have some influence over the price of the selected altcoin. The predictions were made with the use of an artificial neural network, an LSTM network, that was trained on labeled data from 2017. The predictions were then made in intervals of one hour ahead, six hours ahead, and one day ahead through early 2018. The predictions showed that it is possible to make somewhat accurate predictions about the future. The predictions that were made one hour ahead were more accurate than both the six hours ahead predictions and the full day ahead predictions. By comparing the loss rates of the neural networks that were only trained on Ethereum, with the loss rates of the networks that trained on both Bitcoin and Ethereum, is was made clear that training on both cryptocurrencies did not improve the prediction accuracies.