1. Set up a monitor
Choose the identifying fields used to match entities against watchlists. Name is always included by default. Date of birth, location and ID number can be turned on or off for each monitor to control how broad or precise matching is.
2. Add individuals or organizations to the monitor
Each entity is screened when added to the monitor. It is then re-screened automatically as the underlying watchlists change.
3. Generate potential matches
Screening compares each entity against watchlist entries using the fields selected for the monitor. Potential matches are returned with a match score and highlighted details:
- Matched name
- Aliases
- Addresses
- Places of birth
- Dates of birth
- Source lists
4. Review potential matches with AI
Arbiter agents review each potential match using the information provided about the individual or organization. The fields selected for screening determine how potential matches are found. Arbiter can also use other information provided about the individual or organization to confirm or dismiss a match, even if that information was not used during screening.
5. Deliver reviewed results
Each adjudicated match is delivered through the configured notifier with:
- A decision: Potential true positive, false positive, need more information or invalid alert
- A decision rationale: A plain-language explanation of how the decision was reached