An Efficient Hashing-based Ensemble Method for Collaborative Outlier Detection
This work addresses privacy-preserving outlier detection for decentralized participants, but it is incremental as it builds on existing hashing-based ensemble methods.
The paper tackled the problem of efficiently aggregating multiple local detectors in collaborative outlier detection without compromising data privacy or accuracy, and the proposed LSH iTables method outperformed recent ensemble competitors in both centralized and decentralized scenarios across real-world datasets.
In collaborative outlier detection, multiple participants exchange their local detectors trained on decentralized devices without exchanging their own data. A key problem of collaborative outlier detection is efficiently aggregating multiple local detectors to form a global detector without breaching the privacy of participants' data and degrading the detection accuracy. We study locality-sensitive hashing-based ensemble methods to detect collaborative outliers since they are mergeable and compatible with differentially private mechanisms. Our proposed LSH iTables is simple and outperforms recent ensemble competitors on centralized and decentralized scenarios over many real-world data sets.