Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 22 for “"Timeseries data"”.
-
Anomaly Detection in the Vibration of Wind Turbine Blades using Gaussian Process Regression
… turbine simulation tool). In this work, several timeseries data are generated through OpenFAST considering healthy and damaged cases, where the damage is artificially introduced by decreasing the structural stiffness of one blade at three critical locations along the blade span. The goal of the …
-
EVALUATING AND OPTIMIZING UTILITY PLANT OPERATION USING TREND DATA
Trend data have been the main source for utility plant evaluation and optimization. However, the current practice of trend data processing has not been well addressed in previous research as an important part of the workflow. As a consequence, the evaluation and optimization process can fail due to …
-
Development of a Pressure Sensing System Coupled with Deployable Machine Learning Models for Assessing Residual Limb Fit in Lower Limb Prosthetics
… spatial-temporal patterns in multivariate timeseries data, which is processed as an image. The system was designed for edge deployment on the Sony Spresense microcontroller, maintaining a small model size while achieving high accuracy. Results show that the CNN models, particularly those …
-
Housing and labour market interactions
… the model using Bayesian methods and UK timeseries data from 1971Q2 to 2020Q1. It quantifies spillover elasticities, monetary policy parameters, and shock decompositions, shedding light on the effects of major housing and labour policy interventions during the Thatcher era. Counterfactual …
-
Harnessing Machine Learning for Exploring Sensory Processing Across Scales and Modalities
… machine learning tools into exploratory neural data analysis, enabling data-driven hypothesis generation. To this end, I conducted three experiments aimed at better understanding the challenges posed by neural data across different spatial scales and how machine learning can contribute to …
-
Human Action Recognition from Active Acoustics: Physics Modelling for Representation Learning and Inference Using Generative Probabilistic Graphical Models
… modeling and ultimately doing inference from data in multiple modalities where there exists large amounts of low dimensional data complementary to a much smaller set of high dimensional data. In this instance the low dimensional timeseries data are active acoustics from a micro-Doppler sensor …
-
Financial forecasting using artificial neural networks
… scientists have relied only on technical and timeseries data for predicting future prices. So far, the forecasting models have rarely incorporated macro-economic and market fundamentals successfully, especially with short-term predictions ranging less than a month. In this investigation on the …
-
Three essays on the effects of natural resources rents, institutions and demographic factors on economic growth
… for economic growth. Chapter two utilises timeseries data to investigate the impacts of oil rents, institutions, and diversification on the Nigerian economy's growth. Chapter three applies panel data to determine the effects of oil rents and institutions on OPEC economies' growth. Chapter …
-
Classification of Resting-State fMRI using Evolutionary Algorithms: Towards a Brain Imaging Biomarker for Parkinson’s Disease
… connectivity in patients using rs-fMRI data. Specifically, Cartesian Genetic Programming was used to classify dynamic causal modelling data as well as timeseries data. The findings were validated using two other commonly used classification methods (Artificial Neural Networks and Support …
-
peak.gas: An R package for data wrangling and plotting trace gas concentrations from instantaneous output produced by benchtop instruments
… analyzers, provide instantaneous output of timeseries data but may require further processing by the user to estimate injected standard or sample concentrations. Such processing can be time-consuming and prone to error. Here, we developed an open-source package (peak.gas) that integrates …
-
Flexible Models for Heterogeneous Biomedical Data
… development of biomedical sensing techniques and data storage, machine learning has been widely applied to many healthcare applications from the abundance of data resources. However, biomedical data, from real-world applications, has the nature of heterogeneity, and this heterogeneity has not been …
-
Flexible Models for Heterogeneous Biomedical Data
… development of biomedical sensing techniques and data storage, machine learning has been widely applied to many healthcare applications from the abundance of data resources. However, biomedical data, from real-world applications, has the nature of heterogeneity, and this heterogeneity has not been …
-
Investigating the suitability of introducing Index Funds in Kenya
… that adopted a correlational design in which timeseries data from the Nairobi Stock Exchange (NSE) was utilised. To achieve the first objective; the serial (auto) correlation test and the variance ratio tests were employed. The linear unit root tests which comprise of the Augmented …
-
Individualized data generation for electronic health records
… (EHRs) have opened the door to longitudinal data from real-world patients for a wide range of research, such as the creation of clinical decision support systems for different medical applications. Furthermore, the wealth of data collected in EHRs presents a great source for developing …
-
Three essays in macro economics
… and partial adjustment model alongside macro timeseries data, we find pass-through to be moderate, substantial and statistically significant, and our empirical results using an appropriately identified structured VAR model point to moderate and incomplete pass-through irrespective of time …
-
Ανάπτυξη, εφαρμογή και αξιολόγηση σύνθετου δείκτη αναγνώρισης και κατηγοριοποίησης της ξηρασίας και δημιουργία αντίστοιχου συστήματος υποστήριξης αποφάσεων
… central Europe as well. FDI was calculated using timeseries data from 1900-2013, gathered from 170 meteorological stations, distributed in 16 countries throughout the case study, for a variety of temporal steps (3, 6, 9 and 12 months’ time steps). The results were compared to the corresponding …
-
Modelling and measurement of 3d fields in stellarators and tokamaks
… fluctuations. The quantity and complexity of the data obtained from this new array led to the development of a periodic datamining algorithm which has been used to extract distinct plasma fluctuations in multi-channel oscillatory timeseries data. A state of the art synchronous imaging technique …
-
System Identification and Optimal Control of Complex Brain Networks
… geodesic regression model is fit to a dataset involving 30 participants who performed a modified Eriksen Flanker task, to learn within- and between- subject connectivity statistics. Differences in dynamic FC are highly predictive of trial characteristics, as a linear classifier can …
-
Individualized Wrist Motion Models for Detecting Eating Episodes Using Deep Learning
… trained a wrist motion classifier using a large data set collected from 351 people to learn general eating behaviors. We call this a group model. This thesis investigates training the classifier with the same model architecture on new data collected by 8 people, and training the individualized …
-
Essays on financial technologies
… new applications in financial markets, e.g. big data, cloud computing, artificial intelligence, blockchain, cryptocurrencies, peer-to-peer lending, crowdfunding, and robo-advising, inter alia. While traditionally comprising of computer programs and other technology used to support or enable …
Page 1 of 2