The IdentityMind Platform: A Mature Orchestration Hub
January 16, 2020
The recent report by AITE Group: “Fraud, Authentication, and Orchestration Hubs: A Path To Greater Agility”, reinforces the strategic value of the Identity Platform we have built.
We designed our Platform on the premise that the more you know about an individual, the more accurately you are able to assess the risk they represent.
In order to really achieve that value we knew that the following factors were critical:
- A dynamic data structure to represent an identity, based on digital and physical data – this became our patented Electronic DNA (eDNA) technology.
- A way to inform that data structure on behaviors across our client base in a private (hashed) manner.
- A normalization layer on top of an ecosystem of partners of data and technology, that allowed us to tap the best sources of information, and technology; and at the same time provide a unified view regardless of which player in the ecosystem was providing the value.
- A framework enabling an ensemble of risk models that takes advantage of different technologies (e.g. heuristics, machine learning, statistical models) and is informed by the third party ecosystem and by the information collected from our client base.
- Real-time performance, worldwide.
We have written extensively about these functions in our Trusted Digital Identities guide, and also in our KYC Compliance Fundamentals Guide.
Within the Aite framework: “Hub Maturity Curve”, the IdentityMind Platform falls within the “Complex Orchestration Hub” Category, which is defined in the report through the following features:
- One API call to one provider that enables access to multiple detection and/or authentication solutions.
- Risk engine to facilitate scoring driven by machine learning analytics.
- Ability to ingest firm’s internal customer data to contextualize authentication decisions
- Native consortium-data-driven intelligence
Let’s review in detail how the IdentityMind platform meets the above feature set:
One API
“A foundational element of hubs is the ability to access multiple detection and authentication capabilities via a single query to the platform vendor”
Julie Conroy
Aite Group
Fraud, Authentication, and Orchestration Hubs: A Path to Greater Agility
The IdentityMind platform, through a single API, connects our clients to an ecosystem of 20+ partners of data and technology. Through our API you can access the following functions:
- document verification, identity data verification, biometrics, IP geolocation, device intelligence, email reputation, phone risk, virtual currency address risk, and many others
We covered the access and the power of the orchestration API in a recent webinar:
Through our Platform you don’t have to deal with any of the vendors or their business or contractual requirements. You can use any provider in the vendor hub, either automatically or in an ad-hoc fashion. Our partner ecosystem offers services worldwide, and you can customize the platform with the providers you wish by geographic region.
It is important to emphasize that IdentityMind is not a gateway to services. Both the queries and the responses are part of our own data model, our own abstractions. This benefits our clients in two fundamental ways: 1) The platform can use intelligence to route the query to the best provider for a particular function. 2) The interactions with our partners are embedded into our own data model, allowing the platform to understand the semantics of the data and how to use it for risk modeling.
Risk Engine
“True orchestration requires a sophisticated risk engine that can ingest alerts from multiple solutions, assess the transaction risk, and leverage the client’s internal customer data to orchestrate next steps based on the risk scores from the various inputs”
“The risk engine and its associated data ingestion and analytic capabilities are the secret sauce for the hub. This is the cornerstone of the orchestration capabilities and, properly equipped and executed, can enable the platform to tailor risk assessment and authentication at the customer level”
Julie Conroy
Aite Group
Fraud, Authentication, and Orchestration Hubs: A Path to Greater Agility
We know that risk modeling is better achieved through different types of technologies. Machine Learning, without a doubt, is one of them. The platform implements supervised learning to measure the risk of an identity based on its identity graph, or the graph score. The identity graph is a representation of the digital identities being evaluated along with how these identities interact with other identities. This yields a unique vision on the risk of an identity as it is not just the values of their identity attributes, nor whether they are verified or not, but also how these attributes interact with others. You can read more about the value of affiliation and affinity here.
The IdentityMind Identity Proofing Framework is unique in the market, and is the most comprehensive framework to proof an identity and to know the risk you may be undertaking by doing business with that identity. The results of the overall framework are informed by a combination of machine learning models, heuristics, statistical models, and rule engines.
The next step is to enable clients to upload and apply their own machine learning models (PMML).
Context
The success of any risk or compliance platform is how well it integrates into the day-to-day operations. The only way to achieve this is to allow for as much context from the business itself to be incorporated and inform the actions within the platform. While the basic functions of identity proofing are native to the IdentityMind platform, how our clients use them, how they make sure those are implemented seamlessly as part of their day-to-day operations requires information that is particular to each business.
In the report Julie Conroy writes:
“Harnessing internal data to feed analytics is also a top challenge for risk executives”
Julie Conroy
Aite Group
Fraud, Authentication, and Orchestration Hubs: A Path to Greater Agility
This is where the IdentityMind risk and compliance platform shines and clearly separates itself from the pack. It offers the following services to apply business context:
1. Digital Identities. The data structure that represents digital identities (Electronic DNA – eDNA) contains contextual information about the past behaviors of a particular identity, how it has been validated, how it interacts with other identities, and what your operations team have thought about that particular identity in the past. This context is the basis for the Graph Score as discussed previously, and it informs many of the ensemble of risk models available in the platform.
2. Tag Framework. Tags are free-form labels that inform all the operational workflows that are supported in the platform as well as custom machine learning models. Tags informs how money laundering or fraud alerts and cases need to be automatically assigned to your analysts, how specific cases may need to be escalated within the organization and such. Tags can have fraud and risk semantics that can be used to inform custom machine learning models.
3. Open API. Clients can pass any information through the APIs and then use this information within the rule system. So any insights about the business and how this impacts the risk of an identity, or how to evaluate such identity are readily available.
Link to API documentation
Risk Matrix.
The IdentityMind platform also offers a JSON encoded format to freely define an algorithm for modeling risk in the platform and apply the proper rules and policies to evaluate a particular identity. This model is solely dependent on the client’s information, and it is a requirement for some KYC regulatory frameworks.
Native Consortium Data
“Can the hub take advantage of its visibility across multiple client endpoints and derive intelligence that benefits its customer base?”
Julie Conroy
Aite Group
Fraud, Authentication, and Orchestration Hubs: A Path to Greater Agility
The eDNA is able to represent behaviors and identities across all the client base. The IdentityMind database holds over 300 million eDNAs, and a trust score based on these behaviors is available for all our clients to benefit from. eDNAs as we have mentioned before, also informs machine learning based graph score which constitutes one of the key assets of the IdentityMind platform.
eDNAs are completely private. While they allow for matching identities they do not contain any private information that can put the privacy of any of the identities it represents in jeopardy. IdentityMind was built with a security by design mentality.
We have discussed the reliability and accuracy of the IdentityMind trust score and what ROI it can bring into real risk and compliance operations.
A Complex Orchestration Hub
The IdentityMind platform excels at the functions of the orchestration hub because we designed the platform from the very beginning to fulfill these functions. At the time we designed it, these concepts were still new to many people, however over the years, the market has realized that there are real benefits. Specifically, when leveraging a unified framework that can evaluate transactions and clients in the context of the business, that can use its own data to inform better risk modeling, that can simplify the workflows that result from dealing with risk scenarios, and can reduce the cost of vendor overhead. The IdentityMind platform has always offered a single API, with a unique unified concept of a digital identity, that normalizes the functions of identity verification and proofing. This informs all the different risk modeling, the decisions, and the operational workflows required by the enterprise to fully fulfill dealing with their client base with the best opportunity to reduce risk.
Knowing at every turn an accurate risk measure of your clients will allow you to provide the best user experience, as both risk and regulatory requirements can be met at the appropriate level without jeopardizing user experience.
If you have questions about IdentityMind’s functionality, reach out, we’re happy to answer your questions and have a meaningful discussion about how our platform can help you.
