We will use the dataset presented in the below table to develop and use a nave Bayes model for a prediction problem. This dataset relates to a fraud detection scenario in which we would like to build a model that predicts whether loan applications are fraudulent or genuine. There are three categorical descriptive features in this dataset. CREDIT HISTORY captures the credit history of the applicant, and its levels are none (the applicant has no previous loans), paid (the applicant had loans previously and has paid them off), current (the applicant has existing loans and are current in repayments), and arrears (the applicant has existing loans and are in arrears in repayments). The GUARANTOR/COAPPLICANT feature records whether the loan applicant has a guarantor or coapplicant associated with the application. The levels are none, guarantor, and coapplicant. The ACCOMMODATION feature refers to the applicant's current accommodation, and the levels are own (the applicant owns their accommodation), rent (the applicant rents their accommodation), and free (the applicant has free accommodation). The binary target feature, FRAUD, tells us whether the loan application turned out to be fraudulent (true or false). A dataset from a loan application fraud detection domain is given as below. CREDIT HISTORY = paid , GUARANTOR/COAPPLICANT = none ACCOMMODATION = rent Given the above query, calculate the probabilities of FRAUD=true and FRAUD=false for this applicant