Recent advances in artificial neural network methods and modeling have opened new pathways for understanding and predicting the exotic bound state properties in semiconductor nanostructures. In this theoretical and modeling work, we propose a combined artificial neural network-based framework for calculating and investigating bound state effects in confined semiconductor nanocrystals, with emphasis on calculating the exciton effective mass, binding energy, and wavefunction. The model is based on quantum field theory and a nonrelativistic Hamiltonian, enabling the prediction of bound state properties. Unlike conventional perturbative methods, the proposed approach captures confinement effects and multi-exciton interactions in regular semiconductor nanostructures. The study further identifies confinement potential by correlating it with specific physical effects. The obtained theoretical and artificial neural network-based framework provides a true and computationally efficient approach for calculating exciton bound state properties in semiconductor GaAs/〖Al〗_0.3 〖Ga 〗_0.7 As quantum dot. The developed methodology offers reliable numerical and computational tools for investigating confined excitonic systems and may be extended to other low-dimensional semiconductor nanostructures, designing a new generation of optoelectronic, biomedical diagnostics, genomic, and medical analysis tools, and bio-imaging semiconductor nanocrystals technologies, and two-dimensional nanomaterials.