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 242 for “"Data Synthesis"”.
-
DATA SYNTHESIS WITH DIFFERENTIAL PRIVACY
Databases store sensitive information, raising significant privacy concerns when the data is utilized and shared. A canonical approach to address this concern is to generate a synthetic database under differential privacy, which provides rigorous guarantees to safeguard privacy. The synthetic …
-
Privacy-preserving seedbased data synthesis
How can we share sensitive datasets in such a way as to maximize utility while simultaneously safeguarding privacy? This thesis proposes an answer to this question by re-framing the problem of sharing sensitive datasets as a data synthesis task. Specifically, we propose a framework to synthesize …
-
Data synthesis for downstream tasks in autonomous driving
Autonomous vehicles have complex data requirements, and data synthesis has emerged as a solution to alleviate data scarcity. This manuscript presents a set of approaches to tackling this scarcity when training supervised downstream autonomous visions tasks with 4 concepts in mind - realism, …
-
Inertial sensor data synthesis and human motion analysis
… their ability to provide comprehensive kinematic data while being non-invasive. However, designing and deploying IMU systems is challenging, as performance depends on multiple factors, e.g., sensor positioning, processing algorithms, and the need for extensive setup testing. Additionally, the …
-
SEMI-SUPERVISED AND ADVERSARIAL DATA SYNTHESIS FOR LANGUAGE MODELING
… Language Models (LLMs), challenges such as high data demands, computational inefficiency, and vulnerability to adversarial inputs limit their practicality. This thesis enhances the efficiency and robustness of NLP models through a data-centric approach. Key contributions include: (1) a …
-
Detecting Bots using Stream-based System with Data Synthesis
… the continuous supply of representative labeled data, which are expensive and time-consuming to obtain in practice. In this thesis, we build a bot detection system with a data synthesis method to explore detecting bots with limited data to address this problem. We collected the network traffic …
-
ClavaDDPM: Multi-relational Data Synthesis with Cluster-guided Diffusion Models
Recent research in tabular data synthesis has focused on single tables, whereas real-world applications often involve complex data with tens or hundreds of interconnected tables. Previous approaches to synthesizing multi-relational (multi-table) data fall short in two key aspects: scalability for …
-
Data synthesis and classification through machine learning for detecting mild cognitive impairment
… Mild Cognitive Impairment (MCI) from gameplay data of a Serious Game (SG). The foundation of the work is a newly developed SG called WAR Cognitive Assessment Tool (WarCAT). WarCAT is based on the familiar card game WAR and is played on mobile devices (phones, tablets). WarCAT captures players’ …
-
Beyond FAIR: Manifesting marine data synthesis products within the ocean observing system
… (BGC) essential ocean variables (EOV) data determines the success of BGC observations in supporting the development of evidence-based strategies for climate change mitigation, and adaptation, ecosystem health conservation, and sustainable resource management. Acknowledging the positive …
-
Model-data synthesis in terrestrial ecosystem modeling: Inverse analysis and inverse uncertainty analysis.
In this thesis we investigate various inversion approaches for a general type of biogeochemical cycle model that describes the carbon sequestration mechanism of terrestrial ecosystem. We formulate the inverse problems in two approaches---a deterministic inverse approach and a probabilistic inverse …
-
Cross-domain adaptation and geometric data synthesis for near-eye to remote gaze tracking
… with diverse subject appearances, for which data generation and ground-truth annotation have remained a costly process. This dependence on data continues to limit the generalization of gaze models for unseen situations. This work contributes a new data adaption approach of combining the …
-
Analysis of Low-Frequency Climate Variability Through Computational Modeling and Tree-Ring Data Synthesis
… the same approach to historical ERAI reanalysis data and comparing the downscaled storm tracks and intensities to the IBTrACS database. Results suggest small effects are observed in averages over all last millennium eruptions which are non-significant in comparison tothe control. However, for …
-
Stream restoration in the Chesapeake Bay watershed: Data synthesis and analysis of interviews with practitioners
… is one part of the strategy. A comprehensive database of over 4700 stream restoration projects across the watershed was compiled to examine where money is spent, what issues motivate restoration, and what approaches are used. The majority of projects were implemented to restore riparian …
-
Multi-modal Data Synthesis from Organotypic Whole Hemisphere Brain Slice Models to Investigate Cellular-Extracellular Interactions
… a suite of open-source software tools that apply data science and statistical methods to data collected in the brain microenvironment. By developing these tools, we increase the complexity of the analysis that can be done on these datasets while maintaining the reproducibility of the analysis and …
-
Tracking agricultural nitrogen fertilizer use and nitrogen loss pathways across scales through data synthesis and modeling approach
Since the high-yield varieties of crops were adopted in the Green Revolution in the late 1960s, tremendous nitrogen (N) fertilizer associated with agrochemicals, irrigation, and mechanization has been used to boost crop production. The excessive N that is beyond crop demand could be lost to the …
-
PyMouse Lifter : real time 3-D pose estimation for mice with only 2-D annotation via data synthesis
… 3‑D kinematics. Existing open‑source 3‑D mouse datasets for training deep-learning models cover only a narrow range of environments and do not generalize well to various laboratory settings. To overcome these limitations, I introduce PyMouse Lifter, a pipeline that automatically reconstructs 3‑D …
-
Soil health in context: an evaluation of soil health indicator sensitivity, variability, and data synthesis in agricultural systems
… assessment: 1) a lack of soil health indicator data representing the combination of agronomic practices and soil conditions relevant to Minnesota, 2) the need to communicate with non-experts about soil variability, especially as it relates to sampling for soil health assessment, interpreting …
-
DP-Select: Improving Utility and Privacy in Tabular Data Synthesis with Differentially Private Generative Adversarial Networks and Differentially Private Selection
… proposes DP-Select, a novel approach to tabular data synthesis that combines DP-GAN and differentially private selection. We develop a mutual information-based selection method that is flexible and scalable for high-dimensional data and large numbers of marginals while being differentially …
-
Strengthening feedbacks between models and data to improve predictions of the terrestrial carbon sink
… To improve predictions, models and observational data must be better integrated (1) to test the assumptions and hypotheses on which models are based and (2) to quantify the uncertainty around model predictions. Iteratively using data to constrain models and models to inform which data to collect …
Page 1 of 13