FraudHacker - Anomaly detection system for medical insurance claims data
Centre for Medicare and Medicaid Services (CMS) is a governing body in United States that administers the nation's major healthcare program. To detect whether a claim is fraudulent or not, FraudHacker used clustering to perform outlier detection on Medicare claims data from CMS. The data was downloaded in CSV format and loaded directly into PostgreSQL database. FraudHacker extracts data from this database and uses them to perform clustering on it for all the physicians of a perticular specialty (e.g. Neurology) in a perticular state. The clustering used is Hierarchical Density-Based Spatial Clustering of Applications with Noice(HDBSCAN). For each physician the number of fraudulent procedures are tallied and transferred to a second database.
Input variables : State, Specialty of physician, Number of days, Amount charged, Amount paid by medicare
Output Variables : Claim status (fraud / not fraud)
Metrics to Monitor
|
Business
|
:
|
Claims Processed |
$ Saved |
FWA Rate |
FWA by CPT |
FWA by Provider
|
|
Infrastructure
|
:
|
Log Bytes |
Logging/User/IAMPolicy |
Logging/User/VPN |
CPU Utilization |
Memory Usage |
Error Count |
Prediction Count |
Prediction Latencies |
Private Endpoint Prediction Latencies |
Private Endpoint Response Count
|
Visit Model :
github.com