Qlik AutoML

Automatically discovers, creates, and selects relevant features from raw data without manual feature engineering. The system analyzes data relationships and generates derived features that improve model performance.

Evaluates multiple machine learning algorithms and automatically selects the best performing model for a given prediction task. Eliminates the need to manually test different model types.

Processes large batches of records through a trained model to generate predictions at scale. Handles scheduled or on-demand batch scoring jobs.

Provides a user-friendly interface for configuring machine learning models without writing code. Guides users through model setup with visual workflows and forms.

Automatically optimizes model hyperparameters to maximize predictive performance. Eliminates manual trial-and-error tuning of model parameters.

Trains machine learning models on historical data to make predictions on new data. Handles the entire training pipeline from data ingestion to model deployment readiness.

Applies trained models to new data to generate predictions. Produces prediction scores or classifications for individual records or batch datasets.

Automatically calculates and displays model performance metrics including accuracy, precision, recall, and other relevant statistics. Provides visual comparisons across different models.

Embeds predictions directly into Qlik dashboards and visualizations. Enables real-time or scheduled prediction updates within existing Qlik analytics environments.

Analyzes input data for quality issues, missing values, outliers, and data type inconsistencies. Provides recommendations for data cleaning and preparation.

Generates reports showing feature importance and model decision factors. Provides basic interpretability of how the model makes predictions.

Packages trained models into deployable artifacts that can be used for scoring in production environments. Handles model versioning and export.