The Analytical Approach

For this project, my primary goal was to uncover hidden patterns and trends within our cleaned dataset. I employed a descriptive analytical approach to understand the data’s composition and identify the key drivers behind the observed outcomes. To conduct this analysis, I used Power BI.

Key Findings & Insights

After a thorough analysis of the dataset, I was able to uncover a few key insights that have direct business implications.

• Insight #1: Card Exposure & Risk

  21% of cards from the dataset were found on the dark web (1304 out of 6146), indicating a high amount of cards were breached.

  2020 is the most common year in which card pins were changed. This could suggest a wave of compromised accounts during this year.

• Insight #2: Financial Impact

  2.6 million transactions were from cards found on the dark web, with a value of 98.13 million dollars. Highlighting an alarming amount of illegal financial activity.

  The median credit limit is 13,000 dollars, which suggests criminals are targeting mid to high-value accounts.

• Insight #3: Behavioral Impact

  Credit cards make up the largest share of cards on the dark web (42.1%), but debit cards account for the highest transaction value (326.09M), possibly due to easier access to cash or fewer fraud protections.

  Swipe transactions lead the pack with 1.37 million instances, followed by 950,000 chip-based, and 280,000 online transactions. This pattern points to a strong preference among fraudsters for physical card exploitation, such as cloning or skimming.

Final Product: The Interactive Dashboard

Images of the Dashboard

Recommendations

Based on the insights, we could first:

1. Flag the high-risk cards for review. By doing this, we can understand other key characteristics of users who are targeted by card fraud.

2. We can also cross-reference this information with other known data breaches to trace the origin of exposure.

3. Use these characteristics to enhance Fraud Detection Models.