Statistical methods in the NMR spectral analysis

The present review provides a brief summary of the most popular statistical protocols applied in the NMR spectral analysis: machine-learning, neural network, computer assisted structure elucidation, and DP4-based probability methods, which produced a real breakthrough in the interpretation of challenging NMR spectra. Those protocols are a broad family of related machine-learning and artificial neural network statistical treatments, the latter being on the cutting edge of modern statistical science providing an alternative to the state-of-the-art multidimensional NMR experiments, which are often combined with quantum chemical calculations at different levels of theory.