> For the complete documentation index, see [llms.txt](https://docs.fractalworks.io/joule/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.fractalworks.io/joule/components/analytics/ml-inferencing.md).

# ML inferencing

Leverage streaming online predictions to drive insights to action

## Overview

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](https://github.com/jpmml) models within a processing pipeline using a predictive processor.

Learn more about Joule's [ML inferencing architecture here](/joule/components/analytics/ml-inferencing/key-concepts.md).

{% hint style="success" %}
Joule provides the capability to perform **near-real-time predictions** to enable best next action use cases.
{% endhint %}

## Use cases

1. Online customer segmentation
2. Predictive service maintenance
3. Behaviour anomaly detection
4. Best next action
5. Customer conversation scoring

## Available ML inferencing options

<table data-card-size="large" data-view="cards"><thead><tr><th></th><th></th><th data-hidden></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><mark style="color:orange;"><strong>Online predictive analytics</strong></mark></td><td>JPMML prediction processor evaluates event feature vectors in near-real-time </td><td></td><td><a href="/joule/components/analytics/ml-inferencing/online-predictive-analytics.md">Online predictive analytics</a></td></tr><tr><td><mark style="color:orange;"><strong>Feature engineering</strong></mark></td><td>Decorate a feature vector with enriched features specific to the deployed model</td><td></td><td><a href="/joule/components/analytics/ml-inferencing/feature-engineering.md">Feature engineering</a></td></tr><tr><td><mark style="color:orange;"><strong>Model audit</strong></mark></td><td>Monitor model predictions and performance metrics to support explainability, detect drift, and manage retraining</td><td></td><td><a href="/joule/components/analytics/ml-inferencing/model-audit.md">Model audit</a></td></tr><tr><td><mark style="color:orange;"><strong>Model management</strong></mark></td><td>Deploy a retrained model directly in to a running Joule with zero down time</td><td></td><td><a href="/joule/components/analytics/ml-inferencing/model-management.md">Model management</a></td></tr></tbody></table>