Wynyard’s predictive analytics combines data from a variety of sources to generate a model of future events.
Variables for the model can be derived from historical data along with other known indicators and will be weighted for testing and machine learning.
The model can be persistent to run on demand when new relevant data is received. Criteria for detecting changes in data can be encoded using the rules engine. Additionally, in-built triggers and alerts can identify if a situation has or is likely to occur, that warrants further investigation. This is useful for employing preventative interventions. The model should also be able to predict and uncover activity that is insignificant in isolation but if undetected, potentially catastrophic.
Wynyard’s powerful rules engine, utilising the risk model, enables an organisation to refine and modify the model over time. New data sources can be incorporated as well as evaluation mechanisms for measuring accuracy.