Ce diaporama a bien été signalé.
Nous utilisons votre profil LinkedIn et vos données d’activité pour vous proposer des publicités personnalisées et pertinentes. Vous pouvez changer vos préférences de publicités à tout moment.

Graph Gurus Episode 3: Anti Fraud and AML Part 1

24 vues

Publié le

Anti Fraud and AML Part 1

Publié dans : Logiciels
  • Soyez le premier à commenter

  • Soyez le premier à aimer ceci

Graph Gurus Episode 3: Anti Fraud and AML Part 1

  1. 1. Graph Gurus Episode 3 Detecting Fraud and Money Laundering In Real Time with a Graph DB, Part 1
  2. 2. © 2018 TigerGraph. All Rights Reserved Welcome ● Attendees are muted but you can talk to us via Chat in Zoom ● We will have 10 min for Q&A at the end ● Send questions at any time using the Q&A tab in the Zoom menu ● The webinar will be recorded ● A link to the presentation and reproducible steps will be emailed 2 Developer Edition Download https://www.tigergraph.com/developer/
  3. 3. © 2018 TigerGraph. All Rights Reserved Today’s Moderator ● BS in Electrical Engineering and Computer Science from UC Berkeley ● MS in Electrical Engineering from Stanford University ● PhD in Computer Science from Kent State University focused on graph data mining ● 15+ years in tech industry 3 Dr. Victor Lee, Director of Product Management
  4. 4. © 2018 TigerGraph. All Rights Reserved Today’s Guru ● BS & MS in Physics from University of Science and Technology of China (USTC) ● PhD in Quantum Computation from University of California, Merced ● 3-Year TigerGraph Veteran ● Solution Architect, Graph Query Language Designer, Database Core Engineer 4 Dr. Dan Hu, Distinguished AI Research Scientist
  5. 5. © 2018 TigerGraph. All Rights Reserved 5 Real-Time Phone-Based Fraud Detection Massive, Worldwide Problem ● 18 Billion robocalls in US in 2017 (hiya.com) ● Spam/Scam - agile, spoofed numbers Customer: ● 600M subscribers ● 300M calls/day, peak 10K calls/sec ● Need: Real-time detection of various types of phone-based fraud
  6. 6. © 2018 TigerGraph. All Rights Reserved 6 Real-Time Phone Anti-Spam/Scam Detection TigerGraph Solution: Real-time graph-based machine learning and decision system Graph Analytics ● Real-Time Machine Learning ○ 118 graph features per call ○ Retrained periodically with 2M calls ● Real-Time Decisions ○ Call recipient sees alert if ML system says call is suspicious ● In production since Dec 2016 Graph Database ● 600M phone numbers (inside and outside network) ● 15B phone-phone call edges (2 month sliding window) ○ Time ○ Duration ● Real-time graph updates Peak 10K+ calls/sec ○ 118 graph features per phone
  7. 7. © 2018 TigerGraph. All Rights Reserved Examples of Graph Features for Machine Learning 7 Good Phone Features Bad Phone Features (1) Short term call duration (2) Empty stable group (3) No call back phone (4) Many rejected calls (5) Average distance > 3 Empty stable group Many rejected calls Average distance > 3 (1) High call back phone (2) Stable group (3) Long term phone (4) Many in-group connections (5) 3-step friend relation Stable group Many in-group connections Good Phone Features 3-step friend relation /// Good phone Bad phone X X X
  8. 8. © 2018 TigerGraph. All Rights Reserved China Mobile - Detecting Phone-Based Fraud by Analyzing Network or Graph Relationship Features 8 ● Each phone node has a fraud flag, indicating it’s a good phone or a bad phone and what type (scam, harassment, advertisement). ● Run real-time GSQL query for each call: ○ Collect 118 features ○ Compute composite score ○ Update fraud flag ○ Return fraud type
  9. 9. Machine Learning with TigerGraph In Depth China Mobile Anti-Fraud/Scam Detection
  10. 10. © 2018 TigerGraph. All Rights Reserved Phone Fraud Real-Time Detection System phone vertex - fraud flag - expiration time target4 target3 - num of call - total duration - call date list - num of rejection target2 target1 ● 600 Million Vertices ● 15+ Billion Edges ● 300 Million Daily Updatesphone_phone
  11. 11. © 2018 TigerGraph. All Rights Reserved Case 1: Call Type was recently flagged
  12. 12. © 2018 TigerGraph. All Rights Reserved Case 2: Call needs to be classified
  13. 13. © 2018 TigerGraph. All Rights Reserved Machine Learning with TigerGraph Real-time Scoring with Multiple ML models in GSQL • Why TigerGraph? • Fast: Real-time response for both feature collection and scoring. • Efficient: Supports aggregation during traversal, multiple features in one. • Easy: Easy to collect complex features (RDBMS needs multi-join). • GSQL/TigerGraph collects 118 graph features and performs fraud scoring with multiple Machine Learning models in real time. • logistic regression • K-clustering • ML models are trained offline; ML model parameters stored as configuration files for GSQL to use for real-time scoring.
  14. 14. © 2018 TigerGraph. All Rights Reserved China Mobile Machine Learning Model Training • Data labels were obtained from police reports and from online third party sources. • 118 graph features analyzed to build fraud detection model. All features collected by one GSQL query. • Training data’s features collected in GSQL in batch processing and stored as CSV file for future Model Training.
  15. 15. © 2018 TigerGraph. All Rights Reserved Examples of Graph Features for Machine Learning 15 Good Phone Features Bad Phone Features (1) Short term call duration (2) Empty stable group (3) No call back phone (4) Many rejected calls (5) Average distance > 3 Empty stable group Many rejected calls Average distance > 3 (1) High call back phone (2) Stable group (3) Long term phone (4) Many in-group connections (5) 3-step friend relation Stable group Many in-group connections Good Phone Features 3-step friend relation /// Good phone Bad phone X X X
  16. 16. © 2018 TigerGraph. All Rights Reserved Graph Features: Stable Group & InGroup Connection • Stable Group: phones in the target group that have regular calls (stable connection) with source phone • Stable InGroup Connections: phones in the target group that have regular calls (stable connection) among themselves Stable Connection defined as ● Has both Call and Callback ● Num of Call is larger than a given limit ● Total Duration is larger than a given limit
  17. 17. © 2018 TigerGraph. All Rights Reserved Stable Group Pseudocode • Step 1: Starting from the given phone vertex, find its 1-step neighbors. • Step 2: Check if a target has both stable outgoing (phone_phone) and stable incoming edges (phone_phone_reversed). source target4 target3 - num of call - total duration - call date list - num of rejection target2 target1 phone_phone phone_phone phone_phone_reversed Stable Connection defined as ● Has both Call and Callback ● Num of Call is larger than a given limit ● Total Duration is larger than a given limit source
  18. 18. © 2018 TigerGraph. All Rights Reserved Stable InGroup Connections Pseudocode • Step 1: Starting from the given phone vertex, find its 1-step neighbors (target group). • Step 2: For each vertex in the target group, find its 1-step neighbors and check for stable connections. • Step 3: Check the stable target for each vertex in the target group source target4 target3 - num of call - total duration - call date list - num of rejection target2 target1phone_phone phone_phone phone_phone_reversed source Stable Connection defined as ● Has both Call and Callback ● Num of Call is larger than a given limit ● Total Duration is larger than a given limit
  19. 19. © 2018 TigerGraph. All Rights Reserved GSQL DEMO http://192.168.55.50:14240/#/query-editor https://github.com/tigergraph/ecosys/tree/master/guru_scripts /fraud_detection_demo
  20. 20. Q&A Please send your questions via the Q&A menu in Zoom 20
  21. 21. © 2018 TigerGraph. All Rights Reserved Episode 4: Sept 26, 2018 Detecting Fraud and Money Laundering in Real-Time with a Graph DB, Part 2 https://info.tigergraph.com/graph-gurus-4 21 REGISTER FOR MORE WEBINARS AT https://www.tigergraph.com/ webinars-and-events/
  22. 22. © 2018 TigerGraph. All Rights Reserved Additional Resources 22 Compare the Developer Edition and Enterprise Free Trial https://www.tigergraph.com/download/ Guru Scripts https://github.com/tigergraph/ecosys/tree/master/guru_scripts Join our Developer Forum https://groups.google.com/a/opengsql.org/forum/#!forum/gsql-users Take the Developer Survey https://www.tigergraph.com/developer-edition-feedback-survey/ @TigerGraphDB youtube.com/tigergraph facebook.com/TigerGraphDB linkedin.com/company/TigerGraph

×