This section describes the use and training of artificial intelligence (AI), including machine learning (ML) models, in LiveAssurance. For details regarding specific LiveAssurance features, refer to the LiveAssurance product documentation.
Product Features
Certain LiveAssurance features use AI algorithms. These algorithms are in use currently for anomaly detection. The calculations will be performed on physical or virtual servers deployed and controlled by the customer or the partner. The algorithms perform the calculations and generate alerts locally without sending this data for processing to BlueCat.
Anomaly Detection
LiveAssurance (formerly known as Infrastructure Assurance and originally Indeni) v8.4 released anomaly detection. This algorithm is focused on the detection of potential performance anomalies on the devices monitored by LiveAssurance.
The model analyzes each performance metric collected from a network device based on Z-score, which measures how many standard deviations above or below the mean a data point is. When a new data point arrives, a Z-score is calculated. If the Z-score of that data point is greater than or less than 3 an alert is generated.
Model Training
Training processes for anomaly detection algorithms will run on physical or virtual servers deployed by the customer. Both the training data and the resulting model will remain within the customer or partner environment.
Training data is exclusively dedicated to fulfilling detection or forecasting requirements for each customer or partner. There will be no cross-pollination of data, model sharing, or reuse of models between customers or partners.