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 56 for “"Timeseries"”.
-
Representation learning in multi-dimensional clinical timeseries for risk and event prediction
There are major practical and technical barriers to understanding human health, and therefore a need for methods that thrive on large, complex, noisy data. In this work, we present machine learning methods that distill large amounts of heterogeneous health data into latent state representations. …
-
Quantifying pelagic primary production and respiration via an automated in-situ incubation system
… column. The high temporal resolution of the timeseries also enabled the development of Monte-Carlo simulation as a new data analysis technique to calculate DO fluxes, with improved performance in noisy timeseries. Deployment of the incubator was conducted near Ucantena Island, Massachusetts, …
-
Winter mixed-layer development in the central Irminger Sea : the effect of strong, intermittent wind events
… deflected around southern Greenland. A heat flux timeseries for the mooring site was constructed that includes the enhancing influence of the tip jet events. This was used to drive a one-dimensional mixed-layer model, which was able to reproduce the observed mixed-layer deepening in both winters. …
-
The Missing Megawatts Problem: Improving Modelling Practices to Prepare for an Uncertain Future
… I also present a novel outputs-based timeseries clustering method which allows models like GenX to optimize grids using longer timeseries of weather and demand data. Based on my work, I recommend that policymakers, grid operators, and market designers establish rigorous standards …
-
Winter mixed-layer development in the central Irminger Sea : the effect of strong, intermittent wind events
… deflected around southern Greenland. A heat flux timeseries for the mooring site was constructed that includes the enhancing influence of the tip jet events. This was used to drive a one-dimensional mixed-layer model, which was able to reproduce the observed mixed-layer deepening in both winters. …
-
Non-linear dynamics identification using Gaussian process prior models within a Bayesian context.
… to allow Gaussian process to handle large-scale timeseries datasets. Models based on multiple independent Gaussian processes are explored in the thesis. These can be split into two main sections, with common explanatory variable and with different explanatory variables. The two approaches are …
-
Null Models For Cultural And Social Evolution
… on cultural features. Using inference in timeseries of alternative word forms and grammatical constructions, I demonstrate a cultural analog of natural selection on a background of netural evolution. Social evolution, on the other hand, implies selection in a social environment and …
-
Optimal Energy Management of a SAGD Microgrid Participating in a Volatile Electricity Market
… by an EMS forecaster using Bayesian DLM and timeseries ARIMA forecast models. The DLM has the ability to incorporate several exogenous inputs with resultant improvements: accuracies of up to 85% are achieved. The EMS provides optimal economic dispatch for generation and storage units within …
-
Leveraging Structure and Knowledge in Clinical and Biomedical Representation Learning
… methods are needed for electronic health record timeseries data; and (4) asserting global structure in pre-training applications through structure-inducing pre-training.
-
Motif discovery in sequential data
… sequential data such as protein structures and timeseries of arbitrary dimension. As more genomes are sequenced and annotated, the need for automated, computational methods for analyzing biological data is increasing rapidly. In broad terms, the goal of this thesis is to treat sequential data …
-
Earthquakes, groundwater and surface deformation : exploring the poroelastic response to megathrust earthquakes
… It then compares the modeled surface deformation timeseries to the measured deformation at a number of GPS stations. The last chapter walks through the suite of models and data interpolation functions included in the open source toolbox built during this project.
-
Semi-Supervised Deep Learning Frameworks for Transmission-Scale Load Disaggregation and Behind-the-meter Solar Prediction
… solar prediction framework is developed based on Timeseries Dense Encoder (TiDE) algorithm, emphasizing low computational costs and high accuracy for large datasets. This work presents a comprehensive, bottom-up framework for disaggregating transmission-scale load profiles and predicting BTM solar …
-
Machine Learning Applications for Time Series Data: Motor Anomaly Detection and Mean Arterial Blood Pressure Estimation
… ML algorithms for two such applications using timeseries data: (1) TinyML for Anomalous Motor Operation Detection, and (2) Estimation of Mean Arterial Blood Pressure (MAP) from ultrasound measurements. In the first application, we explore different algorithms for detecting anomalous fan motor …
-
A Framework for Generalizing Uncertainty in Mobile Network Traffic Prediction
… leverages domain knowledge to identify unique timeseries sub-behaviors within aggregates of network data. This method produces distributions that are more robust towards changes in the spatio-temporal environment. The ensemble of time-series prediction models trained on these distributions …
-
The effectiveness of literacy professional development provided to a high school faculty
… in their classrooms. I was not able to use the timeseries design because too many of the values for the student data were missing. Therefore, only descriptive statistics were used to answer the central research question and two sub Effectiveness of Professional Development 2 questions, which was …
-
High-Latitude Ionospheric Irregularities Characterized Through Machine Learning Methods
… the input signatures in phase and power in Timeseries Clustering. The observed similarity of the signatures was likely due to steepening spectra during an auroral front.</p> <p>Based on the successful distinction, a hypothesis was formulated: Predominant irregularity mechanisms in the …
-
Human mobility and spatial models for infectious disease
… from influenza-like-illness medical claims timeseries at the scale of 3-digit ZIP codes (ZIPs). Driving distance and time are found to give better gravity model fits than great-circle distance to this data and simulations highlight spatial differences in the spread predicted by the different …
-
Uncovering the Potential of Utilising Terrestrial Biogenic Markers in Ice Cores as Proxies to Past Environmental Conditions
… ice core, producing a complete 1300-year timeseries of all ten target SOA-markers. Through comparison with past land use data, reconstructed temperature records, and known sociological events, it was uncovered that this data can indicate and confirm several characteristics of past …
-
Relationship Between West African monsoon precipitation characteristics and maize yields across Sub-Saharan West Africa
… country-level maize yields that have undergone timeseries analysis to remove trends occurring independently of the WAM. The metrics most correlated with maize yields while maintaining statistically significant slopes were the minimum of total precipitation, standard deviation of the number of …
-
Phasor Measurement Unit (PMU) : based system for event detection on synchronous generators
… Parameter Estimation approach used the method of timeseries analysis for finding the necessary parameters of the generators and modeling them. These parameters were: the inertia of the generators, the speed regulation constant, and the time constant of the turbine-governor system. Axiomatic Design …
Page 1 of 3