Fractals - Digital Payment Fraud Prevention
Fractals the enterprise fraud detection and prevention solution from NCR fraud solution gives you enterprise fraud detection and end-to end fraud competencies across all channels, all accounts and all payment types from a single portal.
With Fractals you can protect transactions across any channel and it tackles any type of transactional fraud, giving you protection against established, developing and emerging fraud types. Fractals gives you multi-organization, multi-hierarchy, multi-channel, and multi-currency fraud prevention and detection. When integrated with your payment engine, it provides accurate, real-time, in-flight, blocking of fraud early in the payment cycle to stop fraud in its tracks, minimize losses, and give your customers a seamless experience.
ENTERPRISE FRAUD PORTAL - FRACTALS
Every day, banks, payment processors and merchants protect their customers, their reputation and their bottom line from fraud using NCR's Fraud solution–Fractals.
Fighting fraud is a continuous undertaking for companies and fraud management needs endless attention. Combine this task with the influx of growing electronic payment types and channels that your customers use and your company faces a perfect storm for the potential of damaging fraud attacks and increasing operational costs for fraud management.
Innovation in payment types and channels is challenged by fraudster innovation that is continuously updated with methods to maximize their profits. Fraud has become an international business and criminal gangs seek out the weakest targets for the easiest and highest returns.
Flexible, Efficient and Reduce Total Cost of Ownership
Taking a unified approach to fraud prevention and management, the Fractals portal supports the detection, analytics and management of fraud across users, accounts, products, and channels. Fractals combines analytical models with user-defined rules to give you an exceptionally high fraud detection rate with a low false positive ratio, helping your company reduce your total cost of fraud management. Fractals is flexible, scalable and easily configurable and is appropriate for companies of any size.
The NCR fraud solution gives you enterprise fraud detection and end-to end fraud competencies across all channels, all accounts and all payment types from a single portal. Your company can monitor transactions and assess historical and aggregated behavior across multiple accounts, multiple channels or groups of accounts.
Fractals enables you to enrich your fraud screening capabilities by incorporating data from the widest possible range of sources. In addition to the usual transactional information, you can include data from other internal systems and fraud-scoring models. Fractals also lets you take advantage of information provided by third parties, so you can validate the device, IP address and geo-location of the payment source.
When deployed with Authentic, NCR’s payment processing platform, or any other payment engine, Fractals allows you to react to potential fraud in real-time, near real-time and batch mode.
NCR FRAUD SOLUTION MODULES
Fraud Integration Hub (FIH)
The Fraud Integration Hub takes payment transactions and event messages, in real time , and sends the relevant elements to your core systems and/or third party data sources to enrich the transaction data or log-on data. This is done prior to running any models or rules, allowing the user to have access to a much deeper set of data for the rules and strategies.
Classification Engine (ACE)
Fractals high fraud detection rate materializes because of Bayesian mathematical models detecting the probability of fraud, executing decisions in real-time as particular patterns arise in their transactional data.
The Adaptive Classification Engine can detect even the most subtle changes in fraud trends, enabling it to identify new fraud types as soon as they are activated and prevent them spreading further. ACE is flexible and is an open framework, giving users control over the fraud detection models, more so than other available solutions. ACE can analyze high-volume, multi-channel transactions and behavior activity with ease and contains strategies that self-adjust as transactions and behavior activity is tagged as fraudulent, which enables the model to learn new fraud patterns as they appear.
By constantly updating themselves, these self-adjusting models enable Fractals to keep up with even the most innovative of fraudsters, while minimizing input the required from your fraud analysts.
Dynamic Rules Engine
The Fractals rules engine enables you to define the rules which are used to score transactions and determine if a transaction should be authorized, declined or reviewed.
The rules engine is a browser-based interface where users define, create, test, evaluate and deploy rules. The graphical rule workbench is designed to make rule writing very straightforward for any level of user, making it easy to express complex and varied rules and providing the ability to update rules rapidly in response to changing threats.
Post-scoring actions allow you to define automatic responses according to the outcome of the Adaptive Classification Engine or rules engine. If ‘in flight’ fraud prevention is needed, Fractals can return the appropriate response to the requesting system. Alternatively, Fractals can deliver a message to your communication partner or system of choice–via web service integration–on any account status change and/or non-financial event changes.
Once the Adaptive Classification Engine or the rules engine has identified a suspicious transaction, Fractals gives your fraud analysts the following support:
Channel specific alert views displays the relevant data your analysts need to for a quick response. Giving the necessary information for each channel at the analysts fingertips allows your organization to improve analyst time efficiency.
Fractals contains a comprehensive data warehouse that is built on the data housed within it.
Fractals has an ETL (extract transform load) script which will take the relevant data and send it to the reporting DB for aggregation and processing. This process allows your financial institution to create Ad-Hoc Reports on multiple aspects of a customer’s data.
FRACTALS AT A GLANCE:
NCR FRAUD SOLUTION - FRACTALS
DIGITAL BANKING FRAUD MODULE
Provides Front-End Fraud Detection Capabilities For Online And Mobile Channels
Today’s fraud solutions have to provide customer-centric and channel agnostic detection. Fraud patterns are evolving and this trend is supported by the adoption of advancements in technology and financial services offered through the online and mobile channels. With activity increasing and the types of devices used consistently changing for the online and mobile channels, your organization is faced with more complexity for protecting and securing yourself and your customer and the risks increase with every interaction of each transaction.
Fraudsters capitalize on the weakest access point, leveraging vulnerabilities to gain access to customer data and financial services for money movement.
NCR’s Digital Banking Fraud Module - Fractals
FRACTALS provides Financial Institutions with front-end real- time, near real-time, or batch detection capabilities around suspicious behavior for activity initiated over an online or mobile channel. FRACTALS analyzes related session data, non-financial event data and payment origination data to assign a risk score of the activity and allow the analyst to make an informed decision on the risk of the activity.
Analyzing every activity processed for its inherent risk, based on a number of factors, including 1.) characteristics of the activity (e.g., IP address, amount), 2.) customer specific profiles (e.g., typical IP address for the customer) and 3.) the inclusion of the recent history of enables a financial institution to make a decision about the level of authentication required based on the risk of the activity initiated from the online or mobile channel.
This approach promotes the financial institution’s risk tolerance strategy effectively while minimizing the friction the customer experiences — all in real time.
Fractals Solution Module Components Fraud Intelligence Hub (FIH)
A central intelligence hub, the FIH easily integrates with your existing internal systems and external data service to enrich the transaction data. The FIH enables you to set up fraud detection rules based on data from a variety of external sources, such as IP intelligence and device reputation.
Adaptive Classificaton Engine (ACE)
A mathematical model, ACE provides financial institutions with a powerful tool for executing decisions in real-time as particular patterns arise in the behavior activity in the online and mobile channels.
Dynamic Rules Engine
Point & click, configurable rules builder that is browser based and enables the financial institution to define the rules used to determine if additional information is needed or if the activity is suspicious.
Channel specific alert view -presenting the relevant data to your analysts for quick responses and efficient investigations. Giving the right information for each channel at the analysts’ fingertips.
Full data warehouse with reporting cubes (OLAP) that allows fully flexible ad-hoc report creation and scheduling.
NCR FRAUD SOLUTION FRACTALS - CARD ISSUER FRAUD MODULE
FRAUD DETECTION FOR DEBIT, CREDIT AND CONTACTLESS CARDS
By 2025, purchase volume for goods and services by cards is expected to reach $54.9 Trillion1 (USD). Transaction volume is increasing and moving to different payment devices and channels. Card fraud is rising, especially card-not-present (CNP) fraud and your financial institution needs a flexible and responsive Fraud Module to protect your reputation and your customers.
NCR’s Issuer Fraud Module - Fractals
Fractals provides financial institutions with protection for accounts, transactions and payment channels in real- time, near real-time, and batch detection capabilities for the detection of fraudulent transactions.
As adoption of the EMV standard re-focuses fraudster activity, Fractals protects card-not present (CNP) and cross-border payments while continuing to minimize losses from counterfeit or stolen cards used in face-to-face transactions. Your financial institution can block fraudulent transactions in-flight and in real time to stop fraud early in the authorization cycle and minimize losses and enhance the customer experience. Fractals is adaptable and agile and uses the combination of machine-learning analytics and user-specified rules to prevent fraud.
Fractals is designed so that you can set up and validate your chosen rules in seconds and provides the ability to set-up multi-level rules to adapt your existing risk policies, and implement new risk policies when you need to.
Fractals Solution Module Components
Fraud Integration Hub (FIH) - A central integration hub, the FIH integrates with your existing internal systems and external data services to enrich the log-on or transaction data. The FIH enables you to set up fraud detection rules based on data from a variety of external sources, such as IP Location and Type.
Adaptive Classificaton Engine (ACE) - ACE provides financial institutions with a powerful tool for executing decisions in real-time as particular patterns arise in the behavior activity in the online and mobile channels using Bayesian statistical techniques.
Dynamic Rules Engine - Point & click, configurable rules builder that is browser based and enables the financial institution to define the rules used to determine if additional information is needed or if the activity is suspicious.
Alert Management - The alert screen provides the necessary channel data to your analysts including relevant data elements, rule information and model scores for rapid and efficient resolution.
Reporting - Full data warehouse with reporting cubes (OLAP) that allows fully flexible ad-hoc report creation and scheduling.
FRACTALS - ENTERPRISE FRAUD ANALYTICS
ADAPTIVE CL ASSIFICATION ENGINE
Companies are continuously challenged to prevent fraud from occurring and are under pressure from their customers to provide a frictionless, easy and secure transacting and purchasing experience across all the channels used.
As it is with every part of the fraud industry, the sooner fraud is prevented the quicker your company will save money, minimize inconvenience to your customers and realize a return on your investment.
The Adaptive Classification Engine (ACE) is an optional module of NCR’s Enterprise Fraud solution, Fractals, that provides automated, intelligent fraud detection by using a combination of Bayesian statistical analysis and proprietary inference techniques from NCR.
Using ACE will enable your company to build your model faster than other traditional methods without compromising detection rates or increasing false positive rates.
ACE is an optional Fractals module that provides automated, intelligent fraud detection by using a combination of Bayesian statistical analysis and proprietary inference techniques from NCR. By applying mathematical models to incoming data elements, ACE identifies unusual and suspicious activity, calculates a probability-based fraud score that indicates the likelihood of fraud, and then triggers an action.
The model is initially set and tuned using your company’s recent historical data and fraud tags and it will identify the unique fraud patterns facing your organization.
The result is a personalized and highly accurate fraud detection model that can be set up in just a few short weeks – giving you the best performance right from the start. ACE contains many models and each model in ACE is based on a series of mathematical algorithms that are targeted at identifying specific fraud patterns or irregularities in behavior. A typical model contains between 20 and
45 separate algorithms and individual models can be established for certain account types or customer segments for even greater granularity.
Once in production, ithe model continues to learn and self-adjust. Not only does this ensure that Fractals can track and react rapidly to changing fraud trends, it frees your fraud analysts to focus on more value-added tasks.
One of the key benefits of ACE compared with traditional Neural Network models is the transparency with which NCR can deliver results to the end user. Unlike a Neural Network which is a black box, ACE scores are delivered with a clear set of reasons and probabilities related to each transaction score. This “white box” approach gives the end-user much more granular information to work from when assessing a transaction or event and when speaking with a customer. More information means better customer service: the key to successful fraud operations.
ACCURATE ANALYSIS FOR ALL BUSINESSES
The initial ACE for your organization is quick to build. After that, model maintenance is both rapid and cost-effective. Companies that deploy ACE can expect to see notable results from the very first event and transaction and high detection rates remain consistent over several years. At the same time, false positive rates remain low.
ACE uses the initial training data set very efficiently, so it can calibrate its models using relatively small data sets. Consequently, ACE can produce customized, cost-effective and accurate detection models for organizations with small or medium-sized customer portfolios, as well as larger institutions.
ACE enables you to adjust the threshold levels for fraud detection to match the alert-handling capacity of your fraud analysts.
THE FRACTALS RULES ENGINE
ACE is normally deployed together with the Fractals Rules Engine, which enables you to define the rules used to score each event and transaction and determine whether it should be authorized, declined or reviewed in Real-Time, Near Real-Time or Batch.
The rules engine is accessed via a browser-based interface. An authorized user can create, edit, test, evaluate and deploy rules without coding or programming. Fractals’ graphical rule workbench makes it easy to express complex and varied rules, and update them rapidly in response to changing threats.
The most successful fraud operations utilize the ACE model and the Fractals Rules engine together layering the defenses against fraud attacks.
ACE AT A GLANCE
The trademark "FRACTALS" and NCR logo are registered in the U.S. Patent and Trademark Office.