Risk Signal: Global Identity Check (formerly Ekyc Verification)

Global Identity Check is a service that produces a quick overall identity risk assessment by validating and linking the User name, address, email, telephone, and IP address against publicly known directories. Global Identity Check assesses the overall risk of an identity and delivers an overall risk score.

The service provides an overall identity risk score which translates to a passed, rejected or warning outcome.

The overall outcome is determined by looking for linkages, validations and behavior of the input elements across two primary sources of information:

Identity graph
Identity network.

Identity Graph

The Identity Graph is used to validate the identity elements.

For example:

Does this email belong to the person?
What type of phone number is this?

The Identity graph is built from over 70 trusted and thoroughly vetted database sources includingTelcos, Government, Agencies, utilities, Voter registration, and Census data.

Identity Network

The Identity Network focuses on how the identity elements are used online.

For example:

When was this email first/last used?
How many times has this address been used?
Why is this phone number associated with 50 IP addresses?

The network analyzes patterns of how information is being used in digital interactions. It is made up of more than 400M global monthly queries that surface usage patterns of identity data.

Use Case: Lightweight Verification for Onboarding

Onboarding at a Financial Institution will require vetting of the provided personal information as part of an accounting opening process. Global Identity Check offers a lightweight check to validate Name, Address, Email and Phone information.

It is often combined with other risk signals like Device and more stringent checks on Name, Address, DOB and SSN. The outcome of these checks often drive the authentication methods given to the customer.

Supported Credentials

The following values can be uploaded as Prepared Data. Alternatively, they can be extracted by an upstream capability in a workflow.

To use Prepared Data with global identity check as a standalone capability use workflow key 10007.

Key	Type	Mandatory	Description
firstName	string	no	First name of the subject.
middleName	string	no	Middle name of the subject.
lastName	string	no	Last name of the subject.
address	object	yes
address.line1	string	no
address.line2	string	no
address.city	string	yes	City of residence as it appears on the Id.
address.postalCode	string	yes	Postal or zip code.
address.subdivision	string	no	City subdivision of residence as it appears on the Id.
address.country	string	yes	Country in ISO-3166-1 Alpha-3 Code or ISO-3166-1 Alpha-2 Code format.
phoneNumber	string	yes	Phone number of the subject in E.164 format. Country code is required.
email	string	no	Optimal to include.
ipAddress	string	no	The IP address entered as a string. Either IPv4 or IPv6 format is accepted.

Example: Prepared Data Body

{
    "firstName": "M***in",
    "lastName": "C***g",
    "email": "m***inc***g@gmail.com",
    "phoneNumber": "+65644***8",
    "ipAddress": "122.23.23.233",
    "address": {
   	 "line1": "153 Rd",
   	 "line2": null,
   	 "city": "Singapore",
   	 "postalCode": "42***1",
   	 "country": "SGP"
    }
}

Example Response

"capabilities": {
	 "globalIdentityCheck": [{
		 "id": "d1c96816-3beb-490d-9243-7e58cc9ed2f1",
		 "credentials": [{
			 "id": "81e9001e-e4d3-4836-8bda-d475330e8224",
			 "category": "DATA"
		 }],
		 "decision": {
			 "type": "REJECTED",
			 "details": {
				 "label": "HIGH_RISK"
			 }
		 },
		 "data": {
			 "emailValid": false,
			 "emailDomainCreationDate": "1995-08-13",
			 "emailMatchToName": "no - match",
			 "emailFirstSeenDays": 4682,
			 "phoneValid": true,
			 "phoneMatchToName": "match",
			 "phoneLineType": "MOBILE",
			 "phoneCarrier": "Singapore Telecommunications Ltd (singtel)",
			 "phoneCountryCode": "SG",
			 "phoneLastSeenDays": 49.0,
			 "phoneEmailFirstSeenDays": 646.0,
			 "phoneMatchToAddress": "country-match",
			 "addressValidityLevel": "valid",
			 "addressMatchToName": "match",
			 "ipRisk": false,
			 "ipRiskScore": 0.951,
			 "ipLastSeenDays": 4.0,
			 "ipGeolocationCountryCode": "US",
			 "ipGeolocationSubdivision": "Oregon",
			 "ipAddressDistance": 8127.0,
			 "identityNetworkScore": 0.934,
			 "identityRiskScore": 469
		 }
	 }]
 }
 }
 }]
 }

Decision Details Labels

Decision Type	Label	Description
PASSED	LOW_RISK
REJECTED	HIGH_RISK
WARNING	MEDIUM_RISK
NOT_EXECUTED	VALIDATION_FAILED
NOT_EXECUTED	NOT_ENOUGH_DATA
NOT_EXECUTED	PRECONDITION_NOT_FULFILLED
NOT_EXECUTED	TECHNICAL_ERROR

Data

Parameter	Type	Description
emailValid	boolean	true false True if the email address is valid.
emailDomainCreationDate	string	Date that the email domain was created in "yyyy-MM-dd" format.
emailMatchToName	string	"not-found" "match" "no-match" The match status between either of the input names and the queried entity.
emailFirstSeenDays	string	null: input email is missing 0: never seen before, high risk 1-90: very high Risk 91-365: neutral risk 366+: low risk
phoneValid	boolean	true false True if the phone number is valid.
phoneMatchToName	string	"not-found" "match" "no-match" The match status between either of the input names and the queried entity.
phoneLineType	string	Null: when phone number is invalid, or line type is unknown (occurs rarely for some international numbers) Landline Fixed-VoIP Mobile Voicemail Toll-free Premium Non-fixed-VoIP Other:
phoneCarrier	string	The company that provides voice and/or data services for the phone number. Carriers are returned at the MVNO level.
phoneCountryCode	string	The ISO-3166 alpha-2 country code associated with the phone number.
phoneLastSeenDays	integer	Count of days since the phone was last observed in the Identity Network. If the phone has not been observed before, last_seen_days will be 0.
phoneEmailFirstSeenDays	integer	Count of days since the combination of phone and email was first observed in the Identity Network. If that combination has not been observed before, first_seen_days will be 0.
phoneMatchToAddress	string	match – Phone location matches input address line 1, address line 2, city, state, and postal code. postal-match – Phone location postal code matches input address postal code. zip4-match – Phone location postal code zip+4 matches input address postal code zip+4. city-state-match – Phone location city and state matches input address city and state. metro-match – Phone location is in the same metro area as input address. country-match – Phone location country matches input address country. no-match – Phone location does not match input address
addressValidityLevel	string	"valid_to_house_number" "missing_address" "invalid" "valid" "valid_to_street" "valid_to_country" "valid_to_city" "Valid_to_house_number_missing_apt" The most granular level to which the address could be validated. Ex. If the address was only valid to the city level (but not to the house level), it would return “valid_to_city”.
addressMatchToName	string	"not-found" "match" "no-match" The match status between either of the input names and the queried entity.
addressValidityLevel	string	The most granular level to which the address could be validated. Ex. If the address was only valid to the city level (but not to the house level), it would return “valid_to_city”. missing_address – An input address was not provided. invalid – The input address is not valid. valid – The input address is valid. valid_to_country – The input address could only be validated to the country level. This means the country of the input address is valid, but the other elements of the input address were unable to be confirmed as valid or invalid. valid_to_city – The input address was validated to the city level. This means the country, state, city, and postal code of the input address are valid, but the street, house number, and subpremise of the input address were unable to be confirmed as valid or invalid. valid_to_street – The input address was validated to the street level. This means the country, state, city, postal code, and street of the input address are valid, but the house number and subpremise of the input address were unable to be confirmed as valid or invalid. valid_to_house_number – The input address was validated to the street and house number level. This means the country, state, city, postal code, street, and house number of the input address are valid, but the subpremise of the input address was unable to be confirmed as valid or invalid. valid_to_house_number_missing_apt – The input address was validated to the street and house number level. This means the country, state, city, postal code, street, and house number of the input address are valid, but the subpremise of the input address was missing and thus unable to be confirmed as valid or invalid.
ipRisk	boolean	true false True if the IP address is considered risky, based on multiple IP data points and velocity calculations.
ipRiskScore	number	Comprehensive risk score associated with an IP address, with a higher score indicating a riskier IP address. A number between 0 and 1 rounded to three decimal places.
ipLastSeenDays	integer	Count of days since the IP address was last observed in the Identity Network. If the IP address has not been observed before, last_seen_days will be 0.
ipGeolocationCountryCode	string	The ISO-3166 alpha-2 country code associated with the geolocation of the IP address.
ipGeolocationSubdivision	string	More granular detail about the IP address location.
identityNetworkScore	Double	The Identity Network Score is a machine learning prediction that provides insight into how risky a digital interaction is based on activity patterns of the identity elements that are being used. Null: when we experience a service timeout, or did not get enough inputs Range: 0.000 - 1.000: <.33: low risk. 33-.8: uncertain >.8: high risk
identityRiskScore	Integer	Null: when we experience a service timeout Range: 0 - 500 <250: low risk 250-350: uncertain >350: high risk

Activity patterns that the Network Score focuses on include the following:

Velocity: how often element(s) are used
Popularity: At how many merchants element(s) are used
Volatility: how often element(s) are used with other elements
Age/maturity: when elements were first/last seen

The Network Score is derived from an Identity Network, which is made up of more than 400M global monthly queries that surface usage patterns of identity data provided by our network of customers.

The Identity Risk Score is a comprehensive risk score calculated in real time that combines authoritative data (match statuses, metadata, linkages) from the Identity Graph as well as usage patterns of elements in the Identity Network

To return a score, we need at least one valid element. For best results, we recommend sending as much of the following information as possible:

Name
Phone
Address
Email Address