Documentation

Neural Network Alert

This artificial-intelligence algorithm is based on the most advanced neural networks used in the field of anomaly detection and models some of the structural and algorithmic properties of the neocortex. It combines and extends approaches used in Sparse distribution memory, Bayesian networks and spatial and temporal clustering algorithms, while using a tree-shaped hierarchy of nodes that is common in neural networks.

 

The trained model thereby predicts the next value based on historical patterns in your data and then measures the difference between the predicted value und the next actual value. The higher the deviation, the higher the degree of anomaly. Additionally, it provides a measure of certainty which is called fluctuation margin. This fluctuation margin tells you how much confidence can be given to the output.