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 16 of 16 for “"matrix factorisation"”.
-
Generalised Bayesian matrix factorisation models
… analysis and related models for probabilistic matrix factorisation are of central importance to the unsupervised analysis of data, with a colourful history more than a century long. Probabilistic models for matrix factorisation allow us to explore the underlying structure in data, and have …
-
Bayesian matrix factorisation: inference, priors, and data integration
… typically observed, and some are very sparse. Matrix factorisation is a popular group of methods that can be used to analyse these matrices. The idea is that each matrix can be decomposed into two or more smaller matrices, such that their product approximates the original one. This …
-
Use and Development of Matrix Factorisation Techniques in the Field of Brain Imaging
Matrix factorisation treats observations as linear combinations of basis vectors together with, possibly, additive noise. Notable techniques in this family are Principal Components Analysis and Independent Components Analysis. Applied to brain images, matrix factorisation provides insight into the …
-
soMLier: A South African Wine Recommender System
… first along with model-based techniques (such as matrix factorisation and neural networks) when applied to the user rating dataset to generate wine recommendations through the ranking of rating predictions. Respectively, these methods are determined to excel at generating lists of relevant wine …
-
Collaborative Genre Tagging
… The first latent factor models made use of matrix factorisation to learn latent factors, but more recent approaches have made use of neural architectures with embedding layers. This master's dissertation outlines collaborative genre tagging (CGT), a transfer learning application of CF that …
-
A recommender system for e-retail
… for this dissertation, namely, implicit matrix factorisation (IMF) and Bayesian personalised ranking (BPR).
-
Enhancing the study of battery degradation using machine learning to get more out of data
… PXRD data. In the third chapter, non-negative matrix factorisation (NNMF) is used to model *in situ* UV-Vis data from an RFB. The models can track the different molecules in solution during cycling and are able to detect both degradation products that had previously been observed, and the …
-
Maximum likelihood parameter estimation in time series models using sequential Monte Carlo
… in Chapter 7 we formulate the nonnegative matrix factorisation problem as MLE in a specific hidden Markov model and propose online EM algorithms using SMC to perform MLE.
-
Chemodynamics of the Galactic Halo
… 7 examines the application of non-negative matrix factorisation to the chemical and kinematic space of stars in the stellar halo. This is done to categorise the halo into in-situ and ex-situ. Lastly, chapter 8 summarises and re-contextualises these papers, and details future work.
-
Field measurements and atmospheric simulation chamber studies of selected biomass burning marker compounds
… via gas-phase reaction with NO 3 . Positive matrix factorisation was used to identify six different source categories for the VOCs which included fresh and atmospherically processed emissions from solid fuel burning, as well as oxidized emissions from local and regional background sources …
-
Mapping the Genomic Context of Mutagenesis
… imprints. A particularly successful approach is matrix-based mutational signature analysis, which identifies prototypical mutation patterns by applying non-negative matrix factorisation to catalogues of single nucleotide variants and other mutation types. However, mutagenesis is a multifaceted …
-
NMR and neutron total scattering studies of silicon-based anode materials for lithium-ion batteries
… component analysis and a variant of non-negative matrix factorisation. With both of these methods, one of the components was found to be associated with high voltages vs. ${Li \vert{} Li^{+}}$ (i.e. a fully discharged anode). This region has seen very little interest by comparison with the low …
-
Combining population-scale single-cell transcriptomics with germline genetics to understand the biology of inflammatory bowel diseases
… differential gene expression, non-negative matrix factorisation, and heritability partitioning. The results highlighted antigen presentation pathways upregulated in epithelial cells from diseased biopsies whilst T cell, myeloid and B cell populations were enriched in IBD risk heritability. …
-
Spatially-resolved multiomic profiling of the human heart in adulthood and development
… to define these niches, which combined unbiased matrix factorisation with expert human structural annotation. In the nodal niches a novel population of nerve growth factor-expressing glial cells was detected, which co-localised with the pacemaker cells. The analysis showed that the …
-
Towards a Fundamental Understanding of Aqueous Organic Redox Flow Batteries – A Study of Degradation, Aggregation, and Reactivity
… is crucial. The combination of non-negative matrix factorisation (NMF), a principal component analysis, and UV-spectroscopy enabled detailed characterisation of the inter- and intramolecular processes. A tendency for DHAQ to aggregate was identified via an *ex-* and *in-situ* UV-Vis …