You are provided a large dataset of prior credit card transa…

You are provided a large dataset of prior credit card transactions containing the transaction amount, location, purchase time and average daily balance for the customer along with a known outcome of whether the purchase was eventually deemed as fraudulent (Yes or No).  Your company would like you to create a machine learning model that would predict the probability of an incoming transaction as potentially fraudulent. If the model shows promise, then it would be used in a real-time system and would need to make its prediction in less than 1 second of processing time. You have a choice between creating a model using k-NN or using the Decision Tree algorithm.  Answer both questions: Which algorithm do you think would make the better choice? Why do you reach this conclusion?

Click the link to open ExPrep to complete FOUR questions in…

Click the link to open ExPrep to complete FOUR questions in Excel.  Enter Password: AUT16 in the first worksheet to see the remaining questions in the Excel file. Excel will save as you go. Timer in ExPrep does not count for time you have already spent in Brightspace. At completion, Click Turnin at the top to Submit the file in ExPrep. Close the browser tab, return to Brightspace to submit the quiz (ignore the warning that question 11 is not completed). Click Link here: ExPrep Grader