Graph Theory Takes on the Opioid Epidemic
summary
Uses Neo4j graph databases, Cypher queries and Jupyter Python notebooks to link prescriptions, providers and pharmacies and detect fraud rings; Neo4j runs community detection, centrality and path algorithms to flag outliers like a pharmacy with 42% of patients living >500 miles away and a doctor whose eight patients died yet kept prescribing. Public datasets (ARCOS) and graph-enhanced ML feed into Cypher on Spark for fast entity resolution and patient 360. Technique breaks real rings (Seattle case, Panama Papers leads to prosecutions) and scales to millisecond decision windows.
AI-written summary · about
chapters (12)
- 0:00Introduction to graph tech
- 2:30Why use graph databases
- 4:29Connected data and value
- 8:36Panama papers and use cases
- 9:31Opioid epidemic context
- 14:05Graph theory fundamentals
- 17:58Identifying anomalies
- 20:26Fraud detection strategies
- 24:10Healthcare identity resolution
- 29:01Graph algorithms
- 34:00Visualizing complex networks
- 38:32Advanced data analysis
related talks
49mModern Graph Databases
26mGraph Databases will change your (freakin') life
17mEnhancing Benefit Adjudication Through Graph Node Embedding, Clustering, and Outlier Detection
28mRole of Knowledge Graphs in Mental Health Diagnosis and Cure
56mGraph Analytic Systems
43mGraphs: Datastructures to Query
45mApplications of Graph in Fraud Detection
05mKeynote Announcement with Fred Trotter