You are developing a model to detect fraudulent credit card transactions. It is crucial to prioritize detection accuracy because missing even a single fraudulent transaction could severely impact the credit card holder. You've used AutoML to train a model based on users' profile information and credit card transaction data. However, after training the initial model, you notice that the model is failing to detect a significant number of fraudulent transactions. To improve the model's performance in accurately identifying fraud, which two training parameters should you adjust in AutoML?