Predictive, pattern based models are built automatically by the Quantum Leap Analyst. There are two types of predictive models:

  • Multi-state classification models for predicting categorical targets.
  • Regression models for predicting continuous valued targets.

  • Make Predictions

    Make a Prediction allows you to interact directly with the model to make predictions. This can be done by choosing a state or value for each of the attributes that define the model. The example used here is intended to illustrate a scenario in which the user is interested in projecting Length of Stay at the time of admission (Length of Stay will drive cost and impact resource and capacity planning).

    Find Predictions

    Quantum Leap Analyst includes a unique feature that will enable you to interrogate the predictive model in search for predicted outcomes that meet certain criteria. This capability is especially useful in situations where the model contains one or more attributes that you control, or in cases where you want to compare predicted outcomes for a given set of possible decisions or actions.