Goal of this vignette | Reference | Shared experimental setup | Run all four variants | Inspecting the Pareto fronts numerically | Visual comparison of the four fronts | One panel per variant | All variants superposed | A unified "ground truth" front | Runtime cost | Picking a best-compromise tree per variant | Three-objective example for every variant | Summary

Overview | Reference | Installation | A first end-to-end example | Step 1. Generate a random bi-objective instance | Step 2. Run the NSGA-II solver (base variant) | Step 3. Inspect the global Pareto front | Step 4. Plot the Pareto front | Step 5. Decode and plot the best-compromise tree | Working with three objectives | Reproducibility | Where to go next

Goal of this vignette | Reference | Shared experimental setup | Run all four variants | Inspecting the Pareto fronts numerically | Visual comparison of the four fronts | One panel per variant | All variants superposed | A unified "ground truth" front | Runtime cost | Picking a best-compromise tree per variant | Three-objective example for every variant | Summary

Overview | Reference | Installation | A first end-to-end example | Step 1. Generate a random bi-objective instance | Step 2. Run the NSGA-II solver (base variant) | Step 3. Inspect the global Pareto front | Step 4. Plot the Pareto front | Step 5. Decode and plot the best-compromise tree | Working with three objectives | Reproducibility | Where to go next

1. Overview | 2. The geneexpr dataset | 3. Preparing the two distance matrices | 4. Baseline run: NSGA-II only | 5. Enhanced run: NSGA-II + Path-Relinking + Pareto Local Search | 6. Visual comparison of the Pareto fronts | 7. Quantitative comparison | 8. Recovering ground truth | 9. Performance notes for version 0.3.0 | 10. References

Introduction | How does MST-kNN clustering works? | Installation instructions | Example on Indo-European languages data set | Load package | Load example data | Performing MST-kNN clustering algorithm | Getting the results | Visualizing clustering | References