Why is this an anomaly? Explaining anomalies using sequential explanations

作者：

Highlights：

• The area under the analyst certainty curve is a better measure compared to the minimum feature prefix.

• For the outlier-based sequential explanations (SE), the particle swarm optimisation search method only outperforms the greedy search methods that make use of the same outlier scoring measure.

• The sample-based SE, support vector machine recursive feature elimination SE, returned the best performing SE overall.

• The best performing outlier and sample-based SEs outperformed the best performing sequential feature explanations (SFE).

摘要

•The area under the analyst certainty curve is a better measure compared to the minimum feature prefix.•For the outlier-based sequential explanations (SE), the particle swarm optimisation search method only outperforms the greedy search methods that make use of the same outlier scoring measure.•The sample-based SE, support vector machine recursive feature elimination SE, returned the best performing SE overall.•The best performing outlier and sample-based SEs outperformed the best performing sequential feature explanations (SFE).

论文关键词：Outlier explanation,Sequential feature explanation,Sequential explanation,Anomaly validation,Explainable AI

论文评审过程：Received 2 March 2020, Revised 7 July 2021, Accepted 3 August 2021, Available online 11 August 2021, Version of Record 15 August 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108227