ANTIFRAUD: WHAT IT IS AND HOW IT WORKS IN THE BANKING SECTOR

[CarderPlanet]

Fraudsters are getting more and more inventive and resorting to various tricks to profit from someone else's expense. Cybercriminals often deal with financial transactions: they issue loans under false names or conduct illegal transactions. To track fraudulent transactions and prevent loss of funds, offers a solution for banks and microfinance organizations - an anti-fraud service. Scoring models based on Big Data will help to detect suspicious activity in advance and protect the funds of bank and MFO clients.

Content

1. What is antifraud and fraud
2. How financial scoring and antifraud are related
3. Big Data is a business solution for bank security
4. Fraud monitoring from Kyivstar

What is antifraud and fraud
Fraud means "scam" in English. These are actions by which attackers fraudulently misappropriate other people's funds. For example, illegal transactions with bank cards, the use of special programs to steal money from accounts, and the purchase of goods in online stores without payment. There are many methods of deception, but now there is an effective countermeasure for them.
Anti-fraud (fraud monitoring), or fraud monitoring, is a system that evaluates the suspicion of financial transactions in terms of fraud. The service informs about the detection of dubious actions and gives recommendations for their further processing. Anti-fraud scoring is effective in risky transactions such as microlending and online lending. The main task of the anti-fraud system for banks is to minimize dangerous situations in the banking sector.

How financial scoring and antifraud are related
Antifraud for payments is based on financial scoring. This is a service that is used in financial institutions to assess the solvency and reliability of the borrower. Before issuing a card, opening an account or taking a loan, the bank's client goes through a mandatory questionnaire. The list of questions includes demographic, social and professional data.
How does the anti-fraud system work in the bank? The program automatically evaluates the client by his answers and displays a scoring score. If the rating is high, the borrower is given a loan. Low is a sign of intruders or dishonest creditors. A similar procedure takes place with transactions that are made by existing users of the bank.

Scoring helps:

• speed up the processing of applications and the evaluation of transactions;
• protect the company from credit fraud;
• to increase the operational activity of the organization;
• reduce the budget for combating fraud;
• create a highly accurate scoring model to predict fraudulent transactions.

For each company, a personal scoring model is created based on the specifics of the organization. Such a system is convenient for both banks and their clients. Scoring helps to quickly fight fraudsters, and users can be sure that the human factor does not affect the execution or refusal of financial transactions.

Big Data is a business solution for bank security
An expert in working with modern methods of big data analytics. The company's specialists have developed tools based on Big Data that help optimize work in the banking sector and counteract fraudsters.
Fraud monitoring is a process that works like a debugged mechanism. Big Data technology quickly analyzes large amounts of data and filters information according to specified criteria. Thanks to machine learning algorithms, bank employees do not need to manually track transactions. At the same time, the effectiveness of fraud monitoring increases, because without the human factor, the number of false alarms and the skipping of real attacks are reduced.
The main advantage of Big Data is the unique information owned by the mobile operator. The company includes data on the use of the SIM card in the anti-fraud system for banks and checks whether high-risk transactions and transactions have been performed using it. A person may have a perfect reputation in the bank's database, but analyzing the SIM card will show different information. Is confident that the way a subscriber uses mobile communications is an indicator of his reliability.

Fraud monitoring
The telecom operator offers several options for analyzing SIM cards as part of antifraud for payments. Checks provide additional protection for bank clients and microfinance organizations from illegal actions.

1. SIM Check service
   Fraudsters can use the customer number, reissue a new SIM card and use it to gain access to online banking. As a result, the user's funds and the bank's reputation will suffer. To prevent this from happening, at the request of the bank, Data-analysts check the identification number of the SIM card (IMSI), which is tied to the card, and not to the number. If the phone number was used to replace the SIM card shortly before the controversial transaction, there is a high probability that fraudsters are acting. Based on the response received, the bank employees block the transaction and begin a detailed check. To minimize financial risks, use SIM Check along with credit scoring.
2. Checking unique SIM cards
   You can find out how often the SIM cards in the customer's phone have changed over a certain period. Bank employees send a request to Kyivstar, and Data-analysts send the number of unique SIM-cards that were used from one phone for a specific period.
   So what is antifraud? Fraud monitoring is a special system capable of detecting cyberattacks, fraud and illegal activities in banking and payment transactions. To more effectively fight cybercriminals, use the automated Big Data technolog. Antifraud in the banking sector will determine the likelihood of fraudulent actions on the part of customers and warn them in time.

 

[Lord777]

How antifraud works
It is generally accepted that buyers-owners of these same cards – suffer from online fraud with plastic cards. But in fact, online stores get no less. Fraudulent transactions, such as fraud, threaten the seller with loss of money, customers, and reputation.

What is a fraud?
Fraud is the conduct of fraudulent transactions, in particular, through the Internet. There are many types of fraud, most of them are aimed at getting hold of a person's bank card data or plastic itself. Among them are phishing (when the cardholder is contacted by "bank employees" by mail or phone with a request to name the card details, or a well-known site is copied by hackers), skimming (copying card data through special devices installed on ATMs), and simply hacker attacks and viruses sent along with spam by email. In this article, we will talk about online fraud that affects online stores-they are the main focus of attention as customers of our processing center.

Online fraud
Fraud in online stores is primarily dangerous for the stores themselves, or TSPS (retail and service enterprises, as they are also called), because it is through them that fraudulent transactions are made and it is from them that they will be asked if the cardholder declares an illegal debit of funds. The classic scheme works as follows: as a result of skimming/phishing or any other illegal actions, the bank card holder unknowingly transfers to the attackers the data of their card that is sufficient to make a purchase in the online store. The attacker makes a purchase and purchases the product / service. The cardholder, after learning about an unauthorized debit, reports the loss of money to the bank that issued the card. In turn, the bank initiates a Chargeback, i.e. a refund of debited funds, and the TSP must return these funds. If the product has already been received by an attacker, then the TPN "hits" three times: it returns money to the cardholder, loses the product for which it has already been paid to the supplier, plus it can earn a fine for missing a fraudulent transaction-up to a complete ban on accepting online payments.

What is antifraud and how it works
Antifraud is a system for monitoring and preventing fraudulent transactions, which checks every payment in real time, running them through dozens, and sometimes hundreds of filters. Anti-fraud mechanisms work in such a way as to track whether there is anything "unusual"in the payment. The task of the system is to check each transaction, find "suspicious" points in it, and make a decision whether to reject the payment or skip it. The anti-fraud system consists of several components: automatic transaction monitoring, which includes many customizable filters, cardholder authentication and card validation mechanisms, and transaction monitoring in "manual" mode for extreme cases.

Such a system is an extremely expensive development that only banks, stores and services – market giants and specialized services (payment aggregators and processing centers that specialize in accepting payments) can afford. That is why most online services and online stores prefer to use the services of third-party contractors to accept payments.

What filters are available
Here we will give examples of filters from the PayOnline processing center – Depending on the system developer, they may be different.

• Validator filters. Example – a validator for bank card details. When entering the card number on the payment form, the system checks it using the Luna algorithm, so that the system can understand that the buyer has not made a mistake, and the card number entered on the payment form is correct.
• Geographical filters. For example, by country of IP addresses. Statistics show that there is a high level of skimming and card compromise in some African countries, and as a result, payments made from these countries are highly likely to be fraudulent.
• Filters-stop lists. Example: Bank card stoplist. If the system receives data from a card that has already been used for payments marked "Fraud", or the cardholder has reported to the issuing bank that its data was compromised, such a card is included in the stop list-the system knows that transactions cannot be skipped on it, as they will turn out to be fraudulent.
• Filters for matching parameters. Example: matching the country of the payer's IP address and the country of the bank card issuer. If the payment is made from a country other than the country where the card was issued, and the cardholder did not notify the bank in advance about their travel, it is likely that the card details were stolen and used by intruders.
• Authorization limit filters. For example, the limit for the amount of one transaction, the number of authorization attempts from one IP address, or from one bank card. To protect both the payer and other participants in the online payment process, there are restrictions on the number and amount of payments made during a day or other period. For some types of business, a particularly large payment, if it turns out to be fraudulent, can significantly hit the profit when refunded.

In total, the system can include hundreds of different filters, and the more a business area is susceptible to fraudulent actions, the more filters are included and the more finely each of them is configured for a specific online store or online service.

What happens if you disable antifraud completely
The store will start skipping fraudulent payments – significantly more than if the anti-fraud filters worked and checked every transaction. If you use 3-D Secure, where the buyer is required to confirm the payment using a one-time password sent via SMS, the online store can minimize losses. However, in case of massive fraudulent transactions, the store may still be disconnected from the payment system. It is enough that the number of fraudulent transactions reaches 1-2% of the number of all payments on the site for a certain period – After that, the acquiring bank can already block payments.

In a situation where 3-D Secure is not used, the situation can be more than deplorable: the conversion rate to successful payments can reach 100%, but losses from such a rash step will be disastrous for the store. However, in the modern market, it is difficult to imagine the situation with disabling all security mechanisms – under such conditions, processing, acquiring banks, and payment systems will refuse to work with the store even at the activation stage.

What happens if you enable all filters
Here the situation is reversed – if all filters are enabled, the percentage of accepted payments may drop significantly. Such protection can simply kill some businesses: for example, if we are talking about the sale of air tickets, the country restriction may negatively affect sales, because the buyer with a Belarusian bank card may be located in Spain and pay for the ticket on the Russian website. Accordingly, when all filters are enabled, we provide a 100% level of security, but significantly reduce the conversion rate to successful payments – If the country of the issuing bank, the seller's site, and the country from which the purchase is made do not match, this is a reason not to miss the payment.

Antifraud and conversion rates
As you can see, the system for monitoring fraudulent transactions requires fine-tuning in order to maintain a high level of security, while not losing most of the profit.

In our company, we have identified several main ways to solve this problem:

• Individual system configuration for the client-specialists analyze the business of an online store, its average receipt, and the geography of the client base, and configure the necessary filters in accordance with the results obtained.
• Providing the user with a mechanism for" manual approval " of transactions. Here, the online store can see for itself which transactions were considered suspicious by the system, and skip them manually. To confirm the identity of the cardholder, the buyer can even request a copy of the passport and bank card with a masked number (the first six and last four digits) and the reverse side with the owner's signature. Based on these documents, the online store can make its own decision on making a payment.
• Providing the partner with the ability to manage part of the system's AntiFraud elements. This decision is made on an individual basis and depends on a number of factors.

In what case can I disable a number of filters in order to maintain a high level of payment conversion? The best way to choose a conversion is if:

• The online store has a high margin and well-organized work with customers in terms of collecting and verifying user data, verifying, confirming and tracking orders;
• Low-risk goods / services that involve a low level of fraud due to their specifics (housing and utilities, city telephone service, home Internet, public services).

Let's take an example: there are two online stores. One sells branded clothing at a 100% mark – up, the other sells airline tickets at a 3% mark-up. In both cases, 100 payments were made, one of which turned out to be fraudulent. Let's assume that the average receipt in both stores is 10,000 rubles. This means that the clothing store will receive a profit of 10,000 rubles from the next sale (due to a high margin), which will cover losses from fraud. Ticket offices receive only 300 rubles from the sale at a 3% mark-up, and to cover the damage of 10,000 rubles, you will need to sell more than 30 tickets. That is why experienced stores with high margins have more opportunities to "weaken" some filters, because with proper work, income from an increased number of successful payments can reduce losses in the event of fraud.

For low-margin businesses, the losses will be more significant – In their case, to increase the conversion rate, it is better not to disable filters, but to switch them to notification mode. The notification mode will mark "suspicious" transactions and allow the online store to manually make a decision on skipping or rejecting a payment. In case of physical delivery of the goods, the courier will be able to ask for an identity document and present the card with which the payment was made before the delivery of the goods. Possible options for insuring your own risks should always be discussed with the payment partner.