Interpreting the results of the person resource

The Person search in BankID AML returns data to be used in the customer due diligence process.

The Person search contains information retrieved from several sources; Data sources of the person resource

If a match is made against a PEP and/or a Sanction-list, match indicators are also returned, showing how certain the data provider is that it is an actual match, and not a false positive (A match produced by the system that, following subsequent investigation, turns out to be another person). There is an indicator; “NameMatchValue” that is related to the name similarity in your query, and if there is a name match (either an exactly name match or partially the same, as a result of a fuzzy match settings in your query) you will also have an indicator telling you the similarity of the Date of birth in your query related to the Date of Birth information listed on the person; “Matchindicator Date of Birth match”

Please note that not all properties will be filled with values for all matches, and there might be null values. This reflects the information that is available about the matched person.

For the Sanction results a list of known addresses is sometimes included in the match (retrieved from the sources). The address list is included in the service, because alternative addresses could be of value to evaluate the risk associated with the person, for example by giving a reason for further investigation. Note that the address retrieved from the National Registry or residential address collected from Posten through Bisnode, will usually be more up to date and of better quality.

A list of url's will often be delivered and this include links to web pages of relevance to the match.

PEP matches

The information provided in a PEP match can reveal a description of the role and the function the person has today or had earlier, that makes /made him/her a PEP, or why the person is regarded a "Relative or Close Associate" (RCA). If the person is no longer active in the position/role that made him/her a PEP the given date from when he/she resigned will be given. These “historical” PEP’s (that is still defined as a PEP) are subject to enhanced customer due diligence measures for at least one year after the PEP leaves the position or function, and sometimes even for a longer period depending on each entity’s own risk-based procedures.

Norwegians in the list can usually be matched on Date of birth and name, and the list has close to 100% coverage.

Sanction lists matches

The source property states which sanction list the match comes from. BankID AML provides, by default, results from two sanction lists: EU_CONSOLIDATED (owned by the EU Commission) and UN_GLOBAL (owned by the UN security council). Other sanction list, such as OFAC and HM Treasury, can be included.

Please note that these Sanction sources very rarely have the SSN and do not that often have a complete Date of birth on the persons sanctioned. Sometimes the same person is listed with several different birth dates and/or just an indication of birth year to give an indication of the person’s age. A list of fields containing information on date, month and/or year of birth is provided in the results (both in the JSON and the PDF report). This information should always be taken into consideration when deciding whether a hit can be defined as a false positive or not.

Be aware of the quality difference in the listings in sanctions list and the PEP- list. In general, a hit against sanction lists requires some more attention around the background of the math indicators than a hit against the pep_edge list. For PEP-matches a match is almost always a match against both name and complete date of birth (since the PEP- list used in BankID AML is close to 100% covered by date of birth for the listed persons), and there is only one registered date of birth per listed person in the PEP list.

Search mode & accuracy

Matching names against PEP and Sanctions is done through a “fuzzy matching mode” in the API. The search result will contain an indicator “NameMatchValue” in which ranges from 0.0 to 1.0. The higher the value is, the better the name similarity for the name in the request and the name in the lists was. The customer can set a limit to how low the NameMatchValue of a search result may be through the usage of a request parameter “nameMatchThreshold”. In general, a “nameMatchThreshold=1.0” means that all the names in your query must match the exact spelling of the names in the list that you screen against. When using “nameMatchThreshold=1.0” you will find that there will be no match even if you have the same Date of birth as a listed person and you have almost the correct/same name. For example:

“Shafi Sultan Mohammad 01.01.1973” will not match at all with;

“Shafi Sultan Mohammed 01.01.1973”

To ensure that small typos, like mentioned above, do not omit hits against listed units, you need to set the “nameMatchThreshold” a bit lower than 1.0. By default, the system is set to 0.85, but this can be changed according to the end user’s own wish. When the threshold is lower than 1.0, it is recommended that you supply more than one keyword (name part) and always combine it with a Date Of Birth to prevent that the number of false positives becomes overwhelming.

For more details, see the attached user guide for interpretation and use of matching-indicators from the Person resource:

Nnin

The national identity number is returned as part of the person resource. If needed, this can be used for the registration of new customers. Please note that the nnin_altsub attribute returned by the ID Token on the OIDC platform, should not be used for this purpose, only for identifying existing customers.

Empty results

If an empty list is returned, this means that a successful search has been conducted and no matches were made.

This does not necessarily mean that there is no risk associated with this customer, and the Merchant is advised to evaluate if further examinations are needed.