In this thesis, the search for the resonant production of a new massive scalar particle X decaying into a new light scalar particle Y and a standard model Higgs boson H through the process X → YH → bƃbƃ is presented. Data from CERN LHC proton-proton collisions at a centre-of-mass energy of 13 TeV are used, collected by the CMS detector in 2016–2018 and amounting to 138 fb^−1 of integrated luminosity. The search is performed in mass ranges of X (0.9–4 TeV) and Y (60–600 GeV) where both the Y and the H are highly Lorentz-boosted. In this kinematic regime, their b quark-antiquark daughter particles are collimated enough to allow the reconstruction of H and Y using single large area jets each. The mass of one of the large-area jets is required to be compatible with that of the Higgs boson, which is 125 GeV. A scan is performed in a two-dimensional plane spanned by the mass of the other jet, associated to Y, and the invariant mass of both large-area jets used to reconstruct X. The results are interpreted in the context of scalar resonances predicted in the next-to-minimal supersymmetric standard model and also in an extension of the standard model with two additional singlet scalar fields. Upper limits are placed on the production cross section of the process as a function of the masses of X and Y in the 0.1–150 fb range. This is the first search for this process using Lorentz-boosted event topologies and significantly extends the constraints on the studied model. A novel calibration method of the ParticleNet algorithm designed to recognize the decays of a massive particle into a pair of b quarks is also presented. The calibration consists of selecting events with a large-area jet with high momentum and measuring the strength of the Z boson peak on a smooth hadronic background. Previously used method performed the calibration on a set of jets originating from gluon fragmentation into bƃ, with special selection applied to make them more akin to the jets originating from the decays of a massive particle, such as Z or H. The presented measurement demonstrates the possibility of a direct calibration of the tagger and provides a validity test for previously used, indirect measurements.