Kuiz berkaitan panduan · Keras Tahap
Triage Amaran AML dan Kuiz Pengurangan Positif Palsu
Utamakan makluman AML tanpa memadamkan aktiviti yang mencurigakan, menggunakan semakan manusia, label berhati-hati, jejak audit dan langkah kualiti.
soalan 1 daripada 8
Bagaimana AML alert triage menyokong pasukan penyiasatan?
Pratonton setiap soalan8 soalan
Jawapan dan penjelasan muncul semasa anda mengambil kuiz.
- Bagaimana AML alert triage menyokong pasukan penyiasatan?
- Decide which account is guilty without analyst review
- Automatically close every low score
- Organize and prioritize alerts for appropriate investigation
- Replace customer due diligence
- Mengapa amaran yang tidak difailkan tidak secara automatik menjadi positif palsu yang disahkan?
- Every alert is required to become a SAR
- Disposition labels are selective and may not reflect true underlying activity
- A model has directly observed criminal intent
- A validated alert cannot be a false positive
- Apakah risiko yang timbul daripada mengurangkan amaran secara terlalu agresif?
- Investigation time always increases
- Customer data becomes less private
- The model stops using transaction data
- Suspicious patterns or new typologies may be missed
- Bagaimana skor model yang rendah harus digunakan dalam aliran kerja risiko?
- One input to review under approved policy, not automatic proof of no risk
- A guarantee to close the case in every institution
- A public explanation of monitoring thresholds
- A reason to delete supporting records
- Which information helps investigators evaluate a prioritized alert?
- Only the final model score
- Only a customer name
- Transaction timeline, source records, linked entities and uncertainty
- A generated narrative without references
- Mengapa mengambil contoh kes yang model triage deprioritizes atau menutup?
- To increase the model's score
- To test for missed suspicious activity and policy gaps
- To remove audit requirements
- To confirm every low score is safe
- What should be logged if automated closure is allowed by approved policy?
- Only the processing time
- The private rule thresholds on a public page
- Nothing after the case is closed
- Policy basis, model version, evidence and decision rationale
- Which metric is more meaningful than simply counting fewer alerts?
- Review quality, missed-case risk, disposition time and customer impact
- The number of model parameters alone
- GPU utilization alone
- Only the total number of closed cases