List at least 5 (five) data points that are required for the analysis and detection of credit card fraud. (3 marks)
Transaction values
A consumer has a particular pattern in purchasing items and the amount they spend towards buying goods and services. Therefore, when their credit card value figure is tallied, the purchases have some relativity. Therefore, if the cardholder purchases something of high value when they are used to purchasing low value, it would raise some red flags.
Delegate your assignment to our experts and they will do the rest.
Time of transaction
Say, for instance, a cardholder is used to make payments using their cards during the day, and a transaction is made at 0100 hrs in the morning. The cardholder is likely to be defrauded.
Location
If a cardholder lives in California and one transaction reflects Canada while the previous and next transactions register, California, in a span of minutes, raises a red flag. The cardholder cannot be in Canada then back to California in a matter of minutes.
Type of product
A cardholder is used to purchasing household supplies like groceries, kitchen supplies, and other items for home use, then all of a sudden, they begin purchasing clothes and perfumes in large quantities is suspicious of fraud.
Shipping Address
If the shipping address is one different from the customer’s address, it could be counted as a suspicious transaction.
Refer to the data table below and identify 3 (three) errors/issues that could impact the accuracy of your findings. (3 marks)
Under the user ID johnp, account number 25671147, the IP address for where the transaction took place is missing. This transaction happened on 3/6/2020. The same problem is apparent with the user ID davidg account number 51422789 on 1/6/2020. The IP address is also missing.
For johnp, his transaction value is also missing on 1/6/2020. These issues could impact the certainty of findings when analyzing the credit card records.
Refer to the data table below and identify 2 (two) anomalies or unexpected behaviors that would lead you to believe the transaction may be suspect. (2 marks)
User ID johnp is used to making his purchases during the day, and his latest normal purchase has been at 1500 hrs. On the same date of 3/6/2020, the user makes three large purchases between 0111 hrs and 0122 hrs for electronics and tools. These are regarded as abnormal behavior and should be cited as suspicious.
User johnp address during his normal purchases is addressed to 1542, Orchid Lane, WA98706, US while two of his suspicious transactions were in-store, and the other simply put down as P.O. Box 1049. The different address raises a red flag for a suspicious transaction.
Ellend makes her transactions in the evening under the address P.O. Box 1322; however, in a transaction dates 2/7/2020, ellend made a purchase for a laptop for an exorbitant price Ar 0005 hrs - another suspicious transaction.
Briefly explain your key takeaway from the provided data visualization chart. (1 mark)
The visualization chart on the number of transactions against their values shows an unusual behavior in purchases, especially for johnp and ellend. They are used to making low-value purchases then all of a sudden their purchases spike in value by going way higher.
Identify the type of analysis that you are performing when you are analyzing historical credit card data to understand what a fraudulent transaction looks like.
The kind of analysis conducted is predictive analysis as the data shown on the tables above makes an analyst expect what the next transaction is going to be or at what time. If the actual happening is different from the predicted outcome then the transaction is suspicious.
