How to identify reshipping scams with Neo4j
4 likes16,510 views
This document discusses how graphs and graph databases can help identify reshipping scams. It provides an example of how transaction and address data can be modeled as a graph and analyzed to detect patterns indicating fraud. Suspicious patterns are visualized to help analysts investigate potential scams, such as a transaction with a billing address in one country but shipping to another. Turning these insights into actions like stopping transactions and banning users can help combat online fraud.
![The pattern is translated in a
graph language.
Looking for the pattern.
MATCH (a:Transaction)-[r*2..3]-(b:City)
WITH a, COUNT(DISTINCT b) AS group_size, COLLECT(DISTINCT b) AS cities
WHERE group_size > 2
RETURN a, cities](https://image.slidesharecdn.com/howtoidentifyreshippingscamswithneo4j-140819024840-phpapp01/85/How-to-identify-reshipping-scams-with-Neo4j-14-320.jpg)
How to identify reshipping scams with Neo4j
- 1. SAS founded in 2013 in Paris | http://linkurio.us | @linkurious How to identify reshipping scams with Neo4j.
- 2. Introduction. CMO >5 years in consulting MSc Political sciences and Competitive Intelligence Jean Villedieu Sébastien Heymann CEO Gephi Founder Phd in Computer Science and Complex Systems Linkurious is a French startup founded in 2013.
- 3. Father Of Father Of Siblings What is a graph ? This is a graph.
- 4. Father Of Father Of Siblings This is a node This is a relationship What is a graph ? / Nodes & relationshipsWhat is a graph : nodes and relationships. A graph is a set of nodes linked by relationships.
- 5. Some of the domains in which our customers use graphs. People, objects, movies, restaurants, music… Suggest new contacts, help discover new music Antennas, servers, phones, people… Diminish network outages Supplier, roads, warehouses, products… Diminish transportation cost, optimize delivery Supply chains Social networks Communications Differents domains where graphs are important.
- 6. The accomplice reships the goods to the fraudster. The fraudster sells the goods for $. He uses the credit cards to buy goods on a website. He has the goods shipped to an accomplice. Reshipping is used to launder money. A scammer steals a few credit cards and wants to turn it into $. Steal credit cards Order goods online Sell the goods What i s reshipping.
- 7. The accomplice is sometimes a victim too.. The accomplice may think he is doing a normal job.
- 8. Reshipping protects the criminals. Why use reshipping. It is worth taking time and efforts to setup a reshipping network if it helps launder money without getting caught.
- 9. Online fraud cost $3.5 billion per year to e-retailers. Source : http://www.internetretailer.com/2013/03/28/online-fraud-costs-e-retailers-35-billion-2012 $3.5 billion The cost of fraud for ecommerce 1/2.
- 10. Merchants pay $3.10 in costs (replacements and fees) for each dollar of fraud losses they incur. The cost of fraud for ecommerce 2/2. Source : http://www.trulioo.com/blog/2014/05/13/social-login-as-an-added-measure-for-e-commerce-fraud/
- 11. How to make sense of complex data. Is it possible for e- retailers to identify reshipping scams?
- 12. What is a graph ? / Nodes & relationshipsA graph data model for the ecommerce orders. date : 11/08/2014 items : Laptop, gifcard amount : $878 IS_USED_FOR Main Street (Street) Address 1 (Address) San Francisco (City) USA (Country) Order 1 (Transaction) 214.77.224.225 (IP_Address) Address 2 (Address) Detroit (City) Folsom Street (Street) Lagos (City) Nigeria (Country) IS_BILLING_ADDRESS IS_SHIPPING_ADDRESS IS_LOCATED_IN IS_LOCATED_IN IS_LOCATED_IN IS_LOCATED_IN IS_LOCATED_IN IS_LOCATED_IN IS_LOCATED_IN IS_LOCATED_IN
- 13. I’m particularly interested in orders where the billing address, the shipping address and the location of the IP address point in different directions ! Can we find these? A fraud expert designs a fraud detection pattern. Designing a fraud detection pattern.
- 14. The pattern is translated in a graph language. Looking for the pattern. MATCH (a:Transaction)-[r*2..3]-(b:City) WITH a, COUNT(DISTINCT b) AS group_size, COLLECT(DISTINCT b) AS cities WHERE group_size > 2 RETURN a, cities
- 15. A graph database handles the data analysis at scale. We use Neo4j to store and analyse the data. ETL Traditional databases. Graph database. The graph databases helps store the data from various sources and analyse it in real-time to identify potential fraud cases.
- 16. Visualization helps analyse the results. ETL API Traditional database. Graph database. Graph visualization. Graph visualization is used to investigate the suspicious cases. Graph visualization solutions like Linkurious help data analysts investigate graph data faster.
- 17. Visualizing suspicious fraud cases. The countries are in dark green, the cities in blue, the IP address in red, the street in orange, the orders (with their date) in light green and the addresses in asparagus.
- 18. It seems a Nigerian scammer is using stolen credit cards on our website. He has enrolled the help of someone from Boston. We should : - stop the transactions they are involved in ; - ban these users and their IP addresses ; - warn the credit card holders ; - contact the authorities ; The analyst turns the alert into actions to stop the fraud. Turning data into action.
- 19. ● This is an illustration of the potential of graphs. In the real world we’d use more advanced techniques (ex : to handle proxies). ● Address analysis has to be combined with other fraud detection mechanisms (ex : credit card verification). ● Only a subset of the fraudulent transactions can be inspected by a human. ● Graph-based fraud detection can be applied in other domains. Graphs are great for fraud detection. Beyond that example.
- 20. You can do it too! Try Linkurious.
- 21. Contact us to discuss your projects at contact@linkurio.us Conclusion
- 22. GraphGist : http://gist.neo4j.org/?6873cf244c0611533029 Blog post on reshipping : https://linkurio.us/reshipping-scams-and-network- visualization Sample dataset : https://www.dropbox.com/s/aca6rhmz7vx2hgp/reshipping% 20scam%20dataset.zip Additional resources.