A significant degree of misclassification of variable stars through the application of machine learning methods to survey data motivates a search for more reliable and accurate machine learning procedures, especially in light of the very large data cubes that will be generated by future surveys and the need for immediate production of accurate, formalised catalogues of variable behaviour to enable science to proceed. In this study, the efficiency of an ensemble machine learning procedure utilising extreme boosting was determined by application to a large sample of data from the OGLE III and IV surveys and from the \textit{Kepler} mission. Through recursive training of classifiers, the study developed a variable star classification workflow which produced an average efficiency determined with the average precision of the model (0.81 for \textit{Kepler} and 0.91 for OGLE) and the $f-score$ of predictions on the test sets. This suggests that extreme boosting can be presented as one of the favourable shallow learning methods in developing a variable star classifier for future large survey projects.

We revisit the calculation of dark matter relic abundances in scalar-tensor gravity using a generic form $A(\varphi_*) = e^{\beta\varphi_*^2/2}$ for the coupling between the scalar field $\varphi_*$ and the metric, for which detailed Big Bang Nucleosynthesis constraints are available. We find that BBN constraints restrict the modified expansion rate in these models to be almost degenerate with the standard expansion history at the time of dark matter decoupling. In this case the maximum level of enhancement of the dark matter relic density was found to be a factor of $\sim 3$, several orders of magnitude below that found in previous investigations.

The relic density of symmetric and asymmetric dark matter in a Gauss-Bonnet (GB) modified Randall-Sundrum (RS) type II braneworld cosmology is investigated. The existing study of symmetric dark matter in a GB braneworld (Okada and Okada, 2009) found that the expansion rate was reduced compared to that in standard General Relativity (GR), thereby delaying particle freeze-out and resulting in relic abundances which are suppressed by up to $\mathcal{O}(10^{-2})$. This is in direct contrast to the behaviour observed in RS braneworlds where the expansion rate is enhanced and the final relic abundance boosted. However, this finding that relic abundances are suppressed in a GB braneworld is based upon a highly contrived situation in which the GB era evolves directly into a standard GR era, rather than passing through a RS era as is the general situation. This collapse of the RS era requires equating the mass scale $m_{\alpha}$ of the GB modification and the mass scale $m_{\sigma}$ of the brane tension. However, if the GB contribution is to be considered as the lowest order correction from string theory to the RS action, we would expect $m_{\alpha} > m_{\sigma}$. We investigate the effect upon the relic abundance of choosing more realistic values for the ratio $\mathcal{R}_{m} \equiv m_{\alpha}/m_{\sigma}$ and find that the relic abundance can be either enhanced or suppressed by more than two orders of magnitude. However, suppression only occurs for a small range of parameter choices and, overwhelmingly, the predominant situation is that of enhancement as we recover the usual Randall-Sundrum type behaviour in the limit $\mathcal{R}_m \gg 1$. We use the latest observational bound $\Omega_{DM}h^2 = 0.1187 \pm 0.0017$ to constrain the various model parameters and briefly discuss the implications for direct/indirect dark matter detection experiments as well as dark matter particle models.

We investigate the effect of a braneworld expansion era on the relic density of asymmetric dark matter. We find that the enhanced expansion rate in the early universe predicted by the Randall-Sundrum II (RSII) model leads to earlier particle freeze-out and an enhanced relic density. This effect has been observed previously by Okada and Seto (2004) for symmetric dark matter models and here we extend their results to the case of asymmetric dark matter. We also discuss the enhanced asymmetric annihilation rate in the braneworld scenario and its implications for indirect detection experiments.

The cosmological significance of the neutralino sector is studied for a class of models in which electroweak symmetry breaking is seeded by a gauge singlet. Extensive use is made of the renormalisation group equations to significantly reduce the parameter space, by deriving analytic expressions for all the supersymmetry-breaking couplings in terms of the universal gaugino mass $m_{1/2}$, the universal scalar mass $m_0$ and the coupling $A$. The composition of the LSP is determined exactly below the W mass, no approximations are made for sfermion masses, and all particle exchanges are considered in calculating the annihilation cross-section; the relic abundance is then obtained by an analytic approximation. We find that in these models, stable neutralinos may make a significant contribution to the dark matter in the universe.