Machine learning, gently
You do not need a statistics or data-science background to use the advanced
Generate sample data
The advanced scenarios read series, events and graphs that already exist. To run any
Data pipeline & lineage
Effort a multi-step pipeline (clean →
Predictive maintenance (anomaly detection)
Effort a model that learns a machine's healthy
Demand forecasting
Effort a lag + calendar feature set and a gradient-
Early warning (predict before it happens)
Effort a forward-looking labelled dataset and a
Asset health scoring
Effort a composite 0–100 health index from
Fraud classification (graph + ML)
Effort graph-derived features from network traversal,
Sequence forecasting with an LSTM
Effort a sliding-window sequence model that learns
LSTM autoencoder anomaly detection
Effort an LSTM autoencoder that learns normal
K-Means — regimes, cohorts & communities
Effort behavioural and graph feature vectors and
Failure prediction with XGBoost
Effort an engineered tabular feature set and a
Soft sensor with Random Forest
Effort a regression model that infers a hard-to-measure
Process monitoring with PCA
Effort a PCA model of normal operation plus the