Using Machine Learning to Fight Fraud
summary
Shows how machine learning is applied to detect fraudsters in a system, covering data collection, feature preparation, and training the algorithm through to the final results and presentation. Describes the end-to-end workflow used by the Business Intelligence team at Hotmart, from dataset gathering to model evaluation and output display.
AI-written summary · about
chapters (9)
- 0:00Fraud-fighting introduction
- 2:59Fraudster profile identification
- 6:16Proposed ML model
- 11:25Precision and recall evaluation
- 16:32Data engineering challenges
- 18:42Results visualization
- 20:12System practical implementation
- 23:19Model maintenance
- 26:11Q&A
related talks
45mAnti-Fraud in Reverse and the Use of ML Methods in It
13mHarnessing Machine Learning to Combat E-commerce Fraud
39mFraud Detection: Using ML to Identify and Manage Fraudulent Activities
16mLeveraging ML to Create a Reliable Anti-Fraud System
37mThe Role of Machine Learning in Today’s Payment Ecosystems
18mThe impact of mathematics, statistics and AI on fraud measurement
52mModern Fraud Prevention Using Deep Learning
31mCatch more potential online fraud faster with Amazon Fraud Detector