DiagSign is a protected invention born in the labs of the French National Research Centre (CNRS) that enables automatic extraction of state of health indicators of an industrial equipment from its physical data.
DiagSign uses a purely mathematical transformation to automatically generate features/indicators from time series that are relevant, highly discriminant, and very adapted to transitional signals.
DiagSign can generate as many features as needed, possibly thousands that are used as input to build Machine Learning-based predictive models.
DiagSign is agnostic to the type of signal nor does it care what equipment is being diagnosed.
Unlike other solutions based on physical indicators or traditional Machine Learning methods, our Automatic Feature Generation Technology enables:
• Quick design (few days) of predictive models
• A high degree of precision and low false alert rates
• The possibility to start adapting the model without previous labelled incidents (Unsupervised learning)