Compare the strengths and weakness of the traditional fraud detection techniques with the machine learning techniques for fraud detection?

The traditional methods when it comes to fraud detection has proven to be quite effective for a much longer period of time compared to the machine learning techniques. This is because it supervised by a trained individual. This means that that person is in charge of manually observing transactions in an effort to discover & apprehend fraudulent behavior. Its strength right now really relies on its longevity in its application.

But with the advent of technology, more and more data are available & transactions happen with every millisecond that passes. Add to that the fact that those who commit fraud have also gotten quite sophisticated in their methods. To be able to cope with the rising need for detecting fraud in real-time, machine learning techniques or unsupervised techniques have been proven to be more useful & more effective in detecting small anomalies within large data sets.

The federal reserve took an expansionary approach during the crisis. this was done by expanding the money supply and boosting liquidity. this can be seen in the fed's actions of lending to banks, purchasing securities, and lowering the federal funds rate in order to lower overall interest rates. the fed's goal was to increase consumer spending and overall liquidity within the system, and they pursued this by expanding the supply of liquid money.