Coinbase Upgrades Anti Fraud System: Integrating Machine Learning and Rule Engine, Shortening Response Time to Hours Odaily Planet Daily News: Coinbase announced that it is optimizing the rule creation process in its anti fraud system by integrating machine learning models and rule engines to achieve more efficient risk management. At the same time, it has proposed a dual track strategy of "the model is responsible for long-term defense, and the rules are responsible for rapid response", and has built a unified framework to form a feedback loop between the two: rules are used to capture new types of fraudulent behavior, and the model is trained in reverse to continuously improve overall defense capabilities. In terms of specific optimization, Coinbase has significantly improved efficiency by restructuring data structures, automating schema evolution, and introducing Notebook based analysis tools, transforming the previously manual rule creation process into data-driven and automated recommendations. Among them, the performance of rule backtesting has been improved by more than 10 times, and the overall response time has been reduced from days to hours. In addition, the new system recommends parameters through machine learning, which helps to reduce the misjudgment rate and minimize the impact on normal users while combating fraud. Coinbase stated that the next step will be to advance event driven automatic rule generation and explore the "one click transformation" of efficient rules into model features, further moving towards an automated risk management system.