bottom

bank-demoBanking & Financial Data Generator

GEDIS Studio can be used to simulate banking transactions from automatic teller machines (ATM), point of sales or bank offices. Those transactions are controlled with profiles describing the behavior of bank's customers in terms of debt or credit operations, timestamps, locations, amount, etc.

This is the perfect tool for testing fraud detection systems, screening filters and any system having to process financial transactions of any type (bank transfer, credit card, check, cash, ...)

This solution is also used to simulate commercial activity of a set of point of sales over a period of time and for various customer profiles.

In fact, the number of usage of that simulator is very large !

Customer's Profiles

There are multiple profiles available in the shared workspace for sample purpose. You can reuse them or write your own. A transaction profile is a XML file with a collection of usage patterns.

Each pattern describes the characteristics of a set of transactions in terms of :

  1. Location of the transactions
  2. Timestamp distribution and duration of the transactions
  3. The cumulated amount of the transaction
  4. The transactions means (Credit card,Check, Cash, ...)
  5. The purpose of the transaction (type of goods buyed,...)

Time distribution

All the transactions in a profile are distributed in a same period of time by the generator. The default period duration is one week and each pattern describes its own time distribution.

You can control the distribution as the day of week. This will allow you to emphasize for example working days or to week end days. You can use dedicated keyword such as "working_days", "any", "week_end", "Sunday", "Monday" ...

You can also control the time (HH:MM:SS) distribution of the transactions in the selected day. For example, you can control that most of the transaction in a pattern occur in the morning during working hours or at lunch time, etc.

Transactions Locations

Each generated transaction is geolocated (Latitude / Longitude) with a real point of sale. Those data are extracted from database of point of interests and reflect actual data, not simulated or generated data.

Within a pattern you can control how a point of sale will be selected for the transaction generated for this pattern. There are two types of parameter that impact on the selected point of sale:

  1. The location of the transaction (Country, City, Area)
  2. The type of transaction (food, supermarket, books, etc.)

Based on this, the generator will select among a subset of points of sale available in our dataset. Please This e-mail address is being protected from spambots. You need JavaScript enabled to view it for any location you may want transaction for.

Transaction Amounts

It is possible to control the total number of transactions in a pattern and also the global amount to be distributed among the transactions. Those parameters may be constant but also random expression allowing to include more variant on the transaction from customers to customers for which the pattern in applied.

For example, assume you write a pattern with 3 transactions for an total amount of $ 100 and apply that pattern to 1,000 customers. Then, each customer will have exactly 3 transactions and for each customer the total of the transaction will be exactly $ 100. But each customer will (probably) have transactions with different individual amount.

You can also write a pattern with between 2 and 10 transaction and for an amount between $ 200 and $ 1,300. This will produce more variant in the generated data and we strongly encourage to do so.