Machine Learning service
Glossary Item Box
The machine learning (lookup value prediction) service uses statistical analysis methods for machine learning based on historical data. For example, a history of customer communications with customer support is considered historical data in Creatio. The message text, the date, and the account category are used. The result is the [Responsible Group] field.
Machine Learning service workflow
The machine learning service consists of the following components:
- ML Service – machine learning web service. The only component enabling external access.
- Python Engine – a service wrapper for open-source machine learning libraries.
- ML Task Scheduler – task scheduler.
- MySQL– MySQL database. You can access it via the standard 3306 port.
The working principles of the machine learning service are presented in Figure 1.
Fig. 1 – Machine learning service workflow
There are two stages of model processing in Creatio: training and prediction.
The prediction model is an algorithm that builds predictions and enables the system to automatically make decisions based on historical data.
Training
The ML model is “trained” at this stage.
Main training steps:
- Establish a session for data transfer and training.
- Sequentially select a portion of data for the model and upload it to the service.
- Request to include a model into a training queue.
- ml-task-scheduler processes the queue.
- python-task-service performs model training and writes the parameters to the database.
- Creatio occasionally queries the service to get the model status.
- Once the model status is set to Done, the model is ready for prediction.
Prediction
The prediction task is performed through a call to the cloud service, indicating the Id of the model instance and the data for the prediction. The result of the service operation is a set of values with prediction probabilities, which is stored in Creatio in the MLPrediction table.
If there is a prediction in the MLPrediction table for a particular entity record, the predicted values for the field are automatically displayed on the edit page.
Machine Learning service scalability
The employment of Docker and Kubernetes makes the machine learning service scalable.
Machine Learning service compatibility with Creatio products
The on-site machine learning service is compatible with all Creatio products of version 7.10 and up. The cloud machine learning service is compatible with all Creatio products of version 7.13.3 and up. To set up the service in an older Creatio version, use the docker image of the corresponding version available on Docker Hub.
Machine Learning service deployment options
Using predictive analysis in Creatio on-site requires additional preliminary setup.
To set up the service, use a server (physical or virtual machine) with Linux or Windows OS installed. Docker software is used for installing the service components. Download the archive containing the configuration files and installation scripts.
We recommend using a Linux-based server for the production environment. You can only use a Windows-based server for the development environment. Contact the support service to receive Docker containers that care compatible with Windows.
See the “Machine learning service setup” article for more details.
See Also
Video tutorial