This study endeavours to probe the impact of vacancies on the nitrogen K-edge fine structure spectrum (Electron Energy Loss Near Edge Structure, ELNES) of face-centred cubic tantalum nitride (TaN), utilizing density functional theory (DFT) simulations. TaN is renowned in the microelectronics industry, serving as diffusion barriers and adhesion layers, and is recognized in the tool industry for its implementation in wear-resistant coatings. Typically, in such applications, the material is deposited as thin films onto a substrate using methods like Physical Vapor Deposition (PVD) or Chemical Vapor Deposition (CVD). It is noteworthy that defects within the crystal structure can dramatically influence the mechanical and electronic properties of the films. Electron Energy Loss Spectroscopy (EELS) is a prevalent tool utilized in transmission electron microscopy (TEM) for identifying crystalline structures. Specifically, the ELNES, representing the part of the EELS spectrum adjacent to the absorption edge of an atom exhibits a high sensitivity to the local atomic environment. This provides valuable insights into defect presence and the electronic structure of the material. This work aspires to elucidate the relationship between nitrogen K-edge spectra predicted by DFT simulations and the parameters of the local environment in simulated structures containing vacancy defects. TaNx structures for simulation were derived from initial 64-atom, defect-free TaN supercells. Varying amounts of Tantalum and Nitrogen vacancies and Schottky defects were introduced to the defect-free structure using the special quasi-random structures (SQS) method. Each structure's geometry was subsequently optimized through full relaxation in the Vienna Ab-initio Simulations Package (VASP) DFT code, ensuring the acquisition of a potential energy surface minimum. Conversely, N K-edge ELNES spectra calculations were performed using the Wien2K DFT code and its TELNES3 package on the relaxed structures, without the introduction of fractional core holes to reduce computational resources. Analyses of the ELNES spectra in this study are based on correlation methods. Their focus was on: relating observed features, both within individual atoms of a structure and between various structures themselves; comparing their respective ELNES intensity distributions for different applied broadening levels, and correlating these with their respective local environments. Modelling of these correlation methods was executed in Python, utilizing several common modules such as numpy, scikit-learn, and scipy. Preliminary findings notably suggest that for the ELNES response at approximately 5.3 eV above the Fermi energy, a robust positive relationship with the fraction of N vacancies in the structure exists concurrently with a negative one with the fraction of Ta vacancies. Moreover, evidence indicates the existence of relative shifts on the energy axis between structures in observed ELNES features.