Load & map data

We have created an easy interface to upload and map data to our generic data model, making it simple to combine data from one or more sources. In the same uploading process, the data is consolidated.
You can also use the One Prediction’s API.
The uploaded and mapped data points are then available to advanced modelling.

Create models

Once your data is mapped and uploaded. Our system will automatically detect what kind of models you will be able to create based on the available data. Here you will go through our model creation wizard, creating the model you want to predict on later. One Prediction will automatically do feature engineering on your data. Feature engineering is the process of using domain knowledge to extract features from raw data via data mining techniques. These features can be used to improve the performance of machine learning algorithms. – including our cognitive match.

Fully automating the creation of advanced machine learning models. Including testing and production deployment.

Compute Predictions

Similar to the model creation wizard, you will go through a prediction creation wizard. Here you will take some business decisions based on the uploaded data and the data earlier trained.

One Prediction will then automatically do model selection, feature set creation and prediction computing.

Scores are exported and ready to use in any of your existing marketing platforms.

Monitor & retrain models

One Prediction also automatically evaluates model performance, tracks performance history, initiates retraining and redeployment. Based on data drift detection.

Data drift is one of the top reasons for which model accuracy degrades over time. For machine learning models, data drift is the change in model input data that leads to model performance degradation. Monitoring data drift helps detect these model performance issues.