Module dbscan 

DBSCAN Anomaly Detection for Episodes

This module implements the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm for detecting anomalous episodes.

DBSCAN is ideal for this use case because:

• It doesn’t require specifying the number of clusters upfront
• It naturally identifies outliers as noise (points not belonging to any cluster)
• It can find clusters of arbitrary shape

Usage

use do_memory_core::patterns::DBSCANAnomalyDetector;

let detector = DBSCANAnomalyDetector::new();
// Use in async context:
// let anomalies = detector.detect_anomalies(&episodes).await;

Integration

The detector is integrated into the learning cycle and can be called during episode completion to identify unusual patterns.

Re-exports

pub use detector::DBSCANAnomalyDetector;

pub use types::Anomaly;

pub use types::AnomalyReason;

pub use types::ClusterCentroid;

pub use types::DBSCANClusterResult;

pub use types::DBSCANConfig;

pub use types::DBSCANStats;

pub use types::EpisodeCluster;

pub use types::FeatureWeights;

Modules

algorithms

DBSCAN Algorithm Implementation

detector

DBSCAN Anomaly Detector

tests

DBSCAN Tests

types

DBSCAN Types