Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
COINS 2014 - Tromso, Norway
Transcript of COINS 2014 - Tromso, Norway
Blekinge Institute of Technology (BTH), Sweden
We developed three different case studies in the area of the social
simulation of financial transactions for fraud detection research:
A new payment system that uses mobile phones to
ease the payments.
A simulationtool that generates realistic scenarios of a retail store based on transactional data from one of the biggest shoe retailers in Scandinavia.
Our approach towards the simulation of bank transactions, payments and transfers between different people and merchants.
of developing these simulators is that it enables us to produce and
share realistic fraud data
with the research community, without exposing potentially sensitive and private information about the actual source.
Fraud Detection Research
We started some time ago to address the problem of developing novel methods for fraud detection, more specifically Anti-Money Laundering for Mobile Money Payments.
Unfortunately for several reasons, including confidentiality, protection of privacy, the law, internal policies and regulations, it is hard if not impossible for an outsideresearcher to get access to detailed financial data.
We deal with the lack of public availablefinancial data, with the idea that if we can not get access to data one goodalternative is to use a simulator to generate financial data for research.
Modeling the financial behavior of individuals is not a simple task.
PaySim: A Mobile Money Payment Simulator
PaySim is based on a company that developed a mobile money platform to transfer money between users using the phone as a sort of electronic wallet.
Our task was to develop an approach that detects suspicious activities that are indicative of money laundering.
This service was only running in a demo mode.
This prevents us from collecting any data that can be used for analysis of possible detection methods.
We modeled and implemented a MABS that uses the schema of the real mobile money service and generates synthetic data following scenarios based on predictions of what could be possible when the real system starts operating.
Using financial synthetic data sets for fraud detection research
We developed three case studies that implement a Multi-Agent Based Simulation model to address the problem of social simulation of financial transactions for fraud detection research.
Produces asimilartype of overall interaction network that we can observe in theoriginal data.
We used the RetSim simulator to investigate two fraud scenarios to see if threshold based detection could keep the risk of fraud at a predetermined set level.
We argue that simple threshold based detection is enough in most of the cases and there is little economic room for other more advanced fraud detection methods that are more costly to implement.
RetSim: A Shoe Store Agent-Based Simulation for Fraud Detection
BankSim is a bank transaction payment simulator based on aggregated data from transactions.
We had no access to this type of financial data until we participated in a contest introduced by a bank in Spain.
They made available aggregated information about payments transactions in Madrid and Barcelona for a 6 months period.
Image Borrowed from: Austrian Research Centers (secoqc.net)
PaySim was a toy example to discover the potential of simulation in this field
Our main results were given using RetSim while studying how effective is a simple threshold detection method
BankSim is richer in data and we are currently developing a simulation that realistically generates financial data
We got data from a Retail Store!
We decided to simulate it!