A-jcodes / SQL-Fraudulent-Transactions-in-the-Bank

Using SQL queries, I crafted complex algorithms to detect suspicious behavior, uncovering hidden connections and irregularities that may indicate fraudulent transactions

Fraud Detection

Overview

The dataset in this case study contains transaction data from a bank. The step field maps a unit of time in the real world, where 1 step is equivalent to 1 hour of time, and the dataset includes a total of 744 steps (30 days simulation). The type field contains information on the type of transaction, including CASH IN, CASH OUT, DEBIT, PAYMENT, and TRANSFER. Other important fields in the dataset include the amount of the transaction in local currency and information on the customer who initiated the transaction nameOrig ), as well as the customer who received the transaction nameDest ). Additionally, the dataset includes information on the initial balance of the customers' accounts ( oldbalanceOrg and oldbalanceDest ) and the new balance after the transaction ( newbalanceOrig and newbalanceDest). One particularly important aspect of this dataset is the presence of fraudulent transactions. The field 'isFraud' indicates whether the transaction was made by fraudulent agents. These agents aim to profit by taking control of customer accounts and transferring funds to another account before cashing out of the system. The isFlaggedFraud field is used to identify illegal attempts to transfer more than 200.000 in a single transaction.

Data Dictionary

1. step-maps a unit of time in the real world. In this case 1 step is 1 hour of time. Total steps 744 (30 days simulation).
2. type-CASH-IN, CASH-OUT, DEBIT, PAYMENT and TRANSFER.
3. amount-amount of the transaction in local currency.
4. nameOrig-customer who started the transaction
5. oldbalanceOrg-initial balance before the transaction
6. newbalanceOrig-new balance after the transaction
7. nameDest-customer who is the recipient of the transaction
8. oldbalanceDest-initial balance recipient before the transaction. Note that there is not information for customers that start with M (Merchants).
9. newbalanceDest-new balance recipient after the transaction. Note that there is not information for customers that start with M (Merchants).
10. isFraud-This is the transactions made by the fraudulent agents inside the simulation. In this specific dataset the fraudulent behaviorof the agents aims to profit by taking control or customers accounts and try to empty the funds by transferring to another account and then cashing out of the system.
11. isFlaggedFraud-The business model aims to control massive transfers from one account to another and flags illegal attempts. An illegal attempt in this dataset is an attempt to transfer more than 200.000 in a single transaction.

Task

The task is to explore the dataset, analyze the patterns and characteristics of legitimate and fraudulent transactions.

1. How many transactions occurred per transaction type?
2. Which Transaction Type has the highest number of Fraudulent Transactions?
3. What is the average fraudulent transaction amount?
4. What is the Maximum fraudulent transaction amount?
5. What is the Minimum fraudulent transaction amount?
6. Who are the Top 10 customers with the highest amount defrauded?
7. How effective is the bank in flagging fraud?
8. Who are the Top 20 Fraudsters

Action

I meticulously designed and executed SQL queries to extract, transform, and load transactional data into a structured format conducive to fraud analysis. Employing various statistical and analytical techniques, I identified patterns indicative of potential fraud, such as unusual transaction amounts, irregular frequency, or suspicious geographic locations. My approach also involved the creation of advanced SQL scripts to flag and categorize transactions based on predefined criteria.

Result

The implementation of the SQL-based fraud detection system yielded remarkable results. Not only did it significantly enhance the identification of fraudulent activities, but it also provided a valuable tool for proactive monitoring and prevention. The project showcased the effectiveness of SQL as a powerful tool for fraud detection within customer transactions, offering real-time insights that contribute to a more secure financial environment.

In conclusion, my fraud detection project using SQL successfully addressed the critical need for a proactive and efficient method to safeguard customer transactions, emphasizing the pivotal role of SQL in mitigating risks and ensuring the integrity of financial systems.