Machine Learning

Leverage streaming online predictions to drive insights to action

For cutting-edge stream-based use cases, the incorporation of machine learning models is integral for next best action processing. Out-of-the-box, Joule offers the capability to seamlessly deploy JPMML models within a processing pipeline using a predictive processor.

Joule provides the capability to perform near-real-time predictions to enable best next action use cases

Example use cases enabled by Joule

  • Online customer segmentation

  • Predictive service maintenance

  • Behaviour anomaly detection

  • Best next action

  • Customer conversation scoring

Online stream prediction architecture

The Joule architecture seamlessly integrates core features such as feature engineering, prediction, model auditing, and model management. Below, we showcase a predictive use case, leveraging the robust features that Joule brings to the forefront.

Available features

Online predictions

The prediction processor evaluates event feature vectors in near-real-time

Feature engineering

Decorate a feature vector with enriched features specific to the deployed model


Monitor and track model performance for prediction explainability and model retraining

Model Management

Deployed models are activity managed using an in-place replacement algorithm

Last updated