Deliverables

1. Implement Distributed Local Outlier Factor method for anomaly detection in ECL for HPCC systems.
2. Test Distributed LOF method on various real-world datasets and compare with other anomaly detection algorithms like Isolated Forest (iForest), DBSCAN, etc. (Also observe performance by comparing with the python implementations for small size datasets).

3)Explore the scope of improvements in Distributed LOF for enhancing its flexibility in anomaly detection for large datasets. 4) Improvise iForest method using LOF for increased efficiency in anomaly detection. 5) Document the results.

Wish list:

1. Contribute an ECL bundle to find KNNs.
2. Contribute an ECL bundle for unsupervised anomaly detection using LOF.