Phone Fraud Detection
Download as pptx, pdf0 likes580 views
This document proposes a hybrid statistical and machine learning solution to detect phone fraud in real time with minimal false positives. It uses statistical analysis and anomaly detection on live streaming phone data to identify anomalous phone numbers. Machine learning with random forests is then used to evaluate additional call features. Graph analysis methods like triangle counting and PageRank are applied to uncover outliers. Confirmed fraudsters are used to train an ensemble machine learning model to progressively improve fraud identification. The system incorporates active learning to enhance detection over time as it remains in use.
Phone Fraud Detection
- 1. 1 Phone Fraudsters in a Haystack Sri Kanajan, Prasad Telekuntla, Mijail Gomez 3rd place in Tata Telecommunications Global Hackathon
- 2. 2 Leaves International Missed Call Unknowingly Calls Premium Number or Manipulative Advertisement $2 BILLION OF LOST REVENUE FROM TELCOM PROVIDERS Example of Phone Fraud
- 3. 3 Motivations Current statistical solutions have low specificity and sensitivity Human fraud analysts have to continually update their heuristic based rules and thresholds Need an adaptive solution that works in real time with minimal false positives
- 4. 4 Statistical Analysis Anomaly Detection Live Streaming Phone Data Hybrid Statistical and Machine Learning Solution Number of Callers/Callee/Cumulative Call Duration Machine Learning (Random Forests) Evaluation of other features in the call log such as answer indicator, area code, pricing Used Hackathon De-identified Phone Log Dataset 16 GB
- 5. 5 Anomaly Detection Through Statistical Analysis # of Unique Callers per Phone Number # of Unique Callees per Phone Number Cumulative Duration of Calls to Specific Phone Numbers ANOMALOUS Phone Numbers!!
- 6. 6 Statistical Analysis Anomaly Detection Machine Learning (Random Forests) Graph Analysis Anomaly Detection Live Streaming Phone Data Predicted Anomalies Hybrid Statistical and Machine Learning Solution
- 7. 7 Fraud Detection Using Graph Metrics Triangle Counting PageRank Others Note: Goal is to uncover the callers that are very different from the large majority
- 8. 8 Using Principal Component Analysis to uncover the outliers in the graph metrics Fraud Detection Using Graph Metrics Possible Fraudsters!
- 9. 9 Statistical Analysis Anomaly Detection Machine Learning (Random Forests) Graph Analysis Anomaly Detection Live Streaming Phone Data Predicted Anomalies Human Observed Fraud Analyst Hybrid Statistical and Machine Learning Solution Possible Fraud
- 10. 10 Human Fraud Analyst Confirmation of Fraudster www.fraud-detector.net Fraud Detection Using Graph Metrics
- 11. 11 Statistical Analysis Anomaly Detection Machine Learning (Random Forests) Graph Analysis Anomaly Detection Live Streaming Phone Data Predicted Anomalies Confirmed Fraudsters Human Observed Fraud Analyst Hybrid Statistical and Machine Learning Solution Possible Fraud
- 12. 12 Ensemble Model Machine Learning and Statistical With labeled data, the classifier can progressively identify patterns beyond the graph metrics (uses all other features in the raw call log) E.g. patterns in area codes or specific pricing plans used by fraudsters Active learning is done online while the system is active. I.e. the longer the system is in use, the better it gets
- 13. 14 Conclusion Possible False Positive Possible Fraudster