This file accompanies Visual Data Analysis notebook being provided to my FinTech Company CFO.
The goal of these analysis tools is to provide information about possible fraud patterns.
Fraudsters sometimes make several small transactions on a credit card in order to avoid being caught. I isolated those transactions for each cardholder in Sql using the groupby function, and counted the number of transactions for each cardholder. Here are those results:
| ccd_owner_id | number_ignored |
|---|---|
| 1 | 10 |
| 2 | 11 |
| 3 | 3 |
| 4 | 16 |
| 5 | 14 |
| 6 | 6 |
| 7 | 18 |
| 8 | 15 |
| 9 | 3 |
| 10 | 20 |
| 11 | 21 |
| 12 | 26 |
| 13 | 19 |
| 14 | 9 |
| 15 | 12 |
| 16 | 19 |
| 17 | 4 |
| 18 | 19 |
| 19 | 22 |
| 20 | 18 |
| 21 | 4 |
| 22 | 7 |
| 23 | 16 |
| 24 | 22 |
| 25 | 16 |
Several of the cards do have a large number of those transactions. The card holders with the most small transactions are # 12, # 19 and # 24. The high quantities of small transactions for those card holders suggests that their cards have been hacked.
The remaining questions in this section are still under investigation.
Regarding the transactions of customers with cardholder ids of 2 and 18, their transaction activity over the months in question is plotted here:
To better contrast the two sets of activity, their plotted together in this diagram:
My conclusion from the data is that cardholder 2 seems to have only regular transactions, while the transaction record of cardholder 18 does show suspicious spikes. I would suggest that all apppropriate measures be taken by cardholder 18.
Regarding the transactions of the customer with cardholder id of 25, the date is provided in a boxplot as seen below:
From this information, it does appear that her card was used for an expensive purchase once in January, and then again in March. Also in April, May, and multiple times in June. I suggest that she definitely take measures to secure that credit card account as soon as possible.
Additional measures for fraud detection - including standard deviation analysis and quartile review - are being pursued for further investigations. Results will be provided as soon as possible.
This Jupyter Lab notebook utilizes the following libraries:
os
Pandas
Numpy
dotenv
JSON
Pathlib
Alpaca Trade API
MCForecastTools
matplotlib
Besides the elements related to those libraries, additional data to be input by the user includes their currency holdings, their stocks and bonds amounts, and their monthly income amounts.
I would like to first acknowledge the guidance and teaching of our FinTech Boot Camp Instructor, Garth Mortensen, our TA, Alejandro Esquivel, and out Student Success Manager, Angelica Baraona. I also found the collective Stack Overflow wisdom essential as ever. Regarding postgreSQL and PGAdmin4 overall I utilized information from codemy and the tech website collective in general. In dealing with a new environment combining PyViz and Sql, the RealPython.com's Vritual Environments Primer was interesting. Finally, the SQL books (SQL Quickstart Guide by Shields, Essential SQL Alchemy by Myers and Copeland and SQL Cookbook by Molinaro were helpful as I continued in the early phases of my learning process. So many things I learned in this assignment, including that once the git add command is done, if you cleverly add items to the folder after that, the git push won't include them.