IOGOD: An interpretable outlier generation-based outlier detector for categorical databases

作者：

Highlights：

• An Interpretable Outlier Generation-based Outlier Detector for Categorical Databases.

• Our proposal outperforms other interpretable outlier detectors.

• Our proposal obtains better results than 14 algorithms on 39 categorical databases.

摘要

•An Interpretable Outlier Generation-based Outlier Detector for Categorical Databases.•Our proposal outperforms other interpretable outlier detectors.•Our proposal obtains better results than 14 algorithms on 39 categorical databases.

论文关键词：Outlier generation,Interpretability,Patterns,One-class classification,Autoencoder,Anomaly detection

论文评审过程：Received 11 August 2021, Revised 8 January 2022, Accepted 17 January 2022, Available online 6 February 2022, Version of Record 10 February 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116570