You can also use OnnxRuntime, in fact it might be the best option among others. On a high level this is what you can do:

1. Generate model in scikit-learn.
2. Convert the model to onnx format
3. Save to disk
4. Load the model in Java via Onnx Runtime
5. Execute inference

Links :

• https://github.com/microsoft/onnxruntime
• https://onnxruntime.ai/docs/get-started/with-java.html