It's a data mining question.i don't understand.

Randy is an insurance adjuster for a national auto insurance company. Using historical insurance claim data, Randy has built an insurance fraud detection model with the help of a data analyst. Applying the model on the validation data set generated the following confusion matrix. A fraudulent insurance claim is a Class 1 case while a nonfraudulent claim is a Class 0 case. Actual Class Predicted Class 1 Predicted Class 0 Class 1 254 10 Class 0 87 649 The confusion matrix was generated based on a 0.50 cut off rate. If the cost of missing a fraudulent insurance claim is much higher than the cost of investigating a potentially fraudulent insurance claim, should the cutoff rate be increase or decreased when classifying claims? Explain your