XBOS Anomaly Detection

What is XBOS?

Cross interaction based outlier score (XBOS) is a cluster-based algorithm for unsupervised anomaly detection. It uses k-means clustering for the first stage, and then calculate cross interaction between clusters as the second stage. Because of this second stage, A small cluster near another large cluster is treated as if that is a middle cluster, so that the data points belong to the cluster is scored ‘not so anomalous’ as a result.

XBOS assumes independence of the features as same as HBOS. XBOS shows very good performance on Kaggle credit card dataset compared to Isolation Forest and HBOS.


XBOS is a really simple algorithm and implemented in just 55 lines of Python code.

Source code is available at GitHub.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s