PHD Computer Science / Διδακτορικές Διατριβές στην Επιστήμη των Υπολογιστών

We present a methodological framework for studying galaxy evolution by utilizing Graph Theory and network analysis tools. We study the evolutionary processes of ultraluminous infrared galaxies (ULIRGs) and quasars and their underlying physical mechanisms, such as star formation and active galactic nucleus (AGN) activity, through the application of graph theoretical analysis tools. We extract, process and analyse mid-infrared spectra of local (z < 0.4) and high-redshift (0.3 < z < 3.0) ULIRGs and quasars between 5 − 38μm through internally developed Python routines, in order to generate relational networks of ULIRGs and quasars (similarity graphs) based on the similarity of their midinfrared spectra. We examine and compare similarity graphs generated using both linear and non-linear supervised classification methods. We also demonstrate how graph clustering algorithms and network analysis tools can be utilized as unsupervised classification techniques under a unified framework for extracting direct and indirect relations between various galaxy properties and evolutionary stages, which provides an alternative methodology to previous works used for classification in galaxy evolution. Furthermore, our methodology compares the output of several graph clustering algorithms in order to identify the best-performing graph theoretical tools for studying galaxy evolution. Additionally, we extract and compare physical features from the mid-IR spectra of ULIRGs and quasars, such as the polycyclic aromatic hydrocarbons (PAHs) emission and silicate depth absorption/emission features, as indicators for the presence of star-forming regions and obscuring dust respectively, in order to determine the underlying physical mechanisms of each evolutionary stage of ULIRGs. We also perform a detailed comparison between the results of the low-redshift (local) and high-redshift samples of ULIRGs and quasars, in order to investigate the evolution of ULIRGs throughout the history of the Universe. Our analysis identifies five types of ULIRGs based on the physical features of their mid-IR spectra, which is quite consistent with the well-established fork classification diagram by providing a higher level classification scheme. The results of our graph theoretical analysis support the evolutionary paradigm of the merger scenario for ULIRGs and showcase noteworthy distinctions in the distribution of ULIRGs and quasars at different evolutionary stages across different redshift ranges.