When designing the architecture for an artificial neural network, there exist a variety of parameters that can be tuned. It is indeed an art in itself to find the right combination for these parameters to achieve the highest accuracy. In this blog post, we are testing the usage of Talos for hyperparameter optimization of a neural network.
In previous blog post about Credit Card Fraud we used an artificial neural network for predicting whether a given transaction is fraudulent, with great results. However, sometimes in problems like these, where we naturally have an imbalanced distribution in our dataset, we might not always have collected data for the minority of the outcomes, in this case, the fraudulent transactions.
Credit card fraud is an ongoing problem in the bank industry. With the contactless payment possibility and e-commerce it is even easier to misuse another person's credit card.