Machine learning covers a variety of methods ranging from simple regression to deep neural networks. Consequently, the range of applications is broad. Unsupervised learning models can be used to discover similarities in the data, while supervised learning models can be used for classification or predictive analysis. The more complicated methods are particularly suited for modelling complex systems with many parameters and non-linear behavior, and are most successful when high quality data is available, preferably in large amounts.
For industrial applications, utilization of machine learning will require some customization, as off-the-shelf solutions are limited.
