Table 13 Expert assessment for determining anomaly for ARL and fuzzy ARL training dataset
From: Anomaly detection in business processes using process mining and fuzzy association rule learning
No | Fraud | Non-fraud |
|---|---|---|
1 | Case that contains skipped activity | Case that does not contain skipped activity |
2 | Case that violates wrong decision | Case that does not violate wrong decision |
3 | Case that violates wrong resource and wrong duty simultaneously | Case that violates wrong resource but does not violates wrong duty or vice versa |
4 | Case that contains an anomaly | Case that does not contain any anomalies |
5 | Case that violates more than one attribute | Case that violates maximum one attribute |
6 | Case with weight of fraud higher than or equal to 0.3 | Case with weight of fraud lower than 0.3 |