%0 Journal Article %T A Machine Learning Approach to Integral Field Unit Spectroscopy Observations: III. Disentangling Multiple Components in H ii regions %+ Université de Montréal (UdeM) %+ Canada-France-Hawaii Telescope Corporation (CFHT) %+ Joseph Louis LAGRANGE (LAGRANGE) %+ Institut d'Astrophysique de Paris (IAP) %+ University of Hawai'i [Hilo] %+ Australian National University (ANU) %+ Universidade Federal de Santa Catarina = Federal University of Santa Catarina [Florianópolis] (UFSC) %+ University of Saint Andrews %A Rhea, Carter %A Rousseau-Nepton, Laurie %A Prunet, Simon %A Hlavacek-Larrondo, Julie %A Martin, R. Pierre, Pierre %A Grasha, Kathryn %A Asari, Natalia Vale %A Bégin, Théophile %A Vigneron, Benjamin %A Prasow-Émond, Myriam %< avec comité de lecture %@ 0004-637X %J The Astrophysical Journal %I American Astronomical Society %V 923 %N 2 %P 169 %8 2021-10-04 %D 2021 %R 10.3847/1538-4357/ac2c66 %K Machine Learning %K ISM %K Galaxies %K Resolved Emission Components %Z Sciences of the Universe [physics]/Astrophysics [astro-ph]/Instrumentation and Methods for Astrophysic [astro-ph.IM]Journal articles %X In the first two papers of this series (Rhea et al. 2020b; Rhea et al. 2021), we demonstrated the dynamism of machine learning applied to optical spectral analysis by using neural networks to extract kinematic parameters and emission-line ratios directly from the spectra observed by the SITELLE instrument located at the Canada-France-Hawai'i Telescope. In this third installment, we develop a framework using a convolutional neural network trained on synthetic spectra to determine the number of line-of-sight components present in the SN3 filter (656-683nm) spectral range of SITELLE. We compare this methodology to standard practice using Bayesian Inference. Our results demonstrate that a neural network approach returns more accurate results and uses less computational resources over a range of spectral resolutions. Furthermore, we apply the network to SITELLE observations of the merging galaxy system NGC2207/IC2163. We find that the closest interacting sector and the central regions of the galaxies are best characterized by two line-of-sight components while the outskirts and spiral arms are well-constrained by a single component. Determining the number of resolvable components is crucial in disentangling different galactic components in merging systems and properly extracting their respective kinematics. %G English %2 https://cnrs.hal.science/hal-03534347/document %2 https://cnrs.hal.science/hal-03534347/file/2110.00569.pdf %L hal-03534347 %U https://cnrs.hal.science/hal-03534347 %~ INSU %~ UNICE %~ CNRS %~ OCA %~ LAGRANGE %~ IAP %~ UNIV-COTEDAZUR %~ SORBONNE-UNIVERSITE %~ SORBONNE-UNIV %~ SU-SCIENCES %~ SU-TI %~ ALLIANCE-SU