Random Forest

Proximity Between Observations

Similarity matrix calculated in Random Forest that counts the number of times two observations end up in the same terminal leaf across all trees in the forest, which can be used for anomaly detection or clustering.

← Tillbaka