1.List at least 5 (five) data points that are required for the analysis and detection of a credit card fraud. (3 marks)

User ID, IP address, transaction date, shipping address, transaction time, transaction value and unit of transaction are data points required for the analysis. 

2.Identify 3 (three) errors/issues that could impact the accuracy of your findings, based on a data table provided. (3 marks)

The missing IP address of davidg’s purchase on 1 June 2020 could lead to the inaccuracy, because we could not be sure whether it was from the same IP or other skeptical IP.

The missing transaction value of johnp’s purchase on 1 June 2020 could impact the analysis, because the loss of transaction value could confuse us if it’s a normal purchase or not. 

The different formats of transaction dates may lead to misunderstanding of the real transaction dates, which may result in inaccurate analysis.

3.Identify 2 (two) anomalies, or unexpected behaviors, that would lead you to believe the transaction may be suspect, based on a data table provided. (2 marks)

The purchased record of ellend on 2 July 2020 is suspect based on the following reasons. First, the IP address were not the same as usual, this might happen due to the different user of the card. Second, the shipping address was changed to the different P.O. box number, which was not the one the user used before. Third, the transaction time was not as usual but at the midnight, which is not the user’s habit based on past records. All the data we got about this transaction are not following the user’s habit.

The three records of johnp’s purchase on 3 June 2020 were also skeptical. Based on past studies, the three records were suspect because all happened in only 11 minutes. Moreover, transaction value and units purchased of those three records are too much higher than johnp’s past transactions. Some abundant purchases happening within a short of time is not only away from the user’s shopping habit, but also skeptical which might owing to credit card fraud. 

4.Briefly explain your key take-away from the provided data visualization chart. (1 mark)

According to the chart we could find out that ellend’s third transaction and johnp’s fourth and fifth transactions are too much higher than their past ones. Unlike davidg’s records, the strings goes far away from their past patterns, so we should study more about these three records which the chart tells us that is anomalies.

5.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. [Hint: The four types of Analytics include: Descriptive, Diagnostic, Predictive, Prescriptive] (1 mark)

In the beginning, I use the descriptive analysis to point out the the anomalies from all the records provided. I found out that some records are not following its old patterns, which are far away from it, according to the data visualization chart. Then I used diagnostic analysis to dig deeper, I noticed that some important data points are skeptical based on the past studies. Those data points are not like those user’s shopping habit, which we could consider them abnormal, and likely to be a credit card fraud.