Data Mining & Tax Evasion

by Chris S
The article I chose touches on how tax authorities are challenged with collecting and identifying businesses that do not pay their correct taxes. Believe it or not, businesses and everyday people cheat on their taxes, whether it be for a higher refund or to elude from paying more money back to the government. Of all agencies, the IRS is stated to have limited amounts of resources when it comes to detecting fraud. To try and combat this, a screening system using data mining techniques was created in order to try and “value-added tax” reports, or any other type of suspicious activity. Once these particular activities are recognized, then they can be further audited and have proven effective.

I believe this relates to our class in a number of ways. Obviously the government is using a vast database to try and recognize the fraudulent activities and stop them. Data mining is basically analyzing data from a variety of different approaches and summarizing the findings into a useful set of information. If anyone were to get into the database field, chances are there will be situations where people will have to do these kinds of tasks.

The whole data mining subject reminded me of another story I came across about how a grocery store used data mining to their advantage. It may not be 100% accurate, but from what I remember was that it turned out that on weekends, the sale of diapers and beer were above average. Apparently on the weekends, the father would more often be sent to the store to buy some diapers, and occasionally buy some beer as well. After noticing this type of behavior, the grocery stores relocated the diaper section closer to the alcohol section, thus creating more sales.


Wu,¬†Roung-Shiunn, C.S Ou, Hui-ying Lin, She-I Chang, and David Yen. “Using Data Mining Technique to Enhance Tax Evasion Detection Performance.”¬†Expert Systems with Applications, 39.10 (2012): 8769-8777.

3 thoughts on “Data Mining & Tax Evasion”

  1. I agree that businesses are depending on data mining. If they want to have competitive advantage they need to implement data mining to sell better their stuff. The beer and diapers example is a good one because it shows how businesses are using these tool to strategically allocate their inventories to maximize the sales. The IRS is using these tools too to find fraudulent activities.

  2. Databases are a storage for a huge amount of data. Data by itself does nothing when it is just data, but if it is processed, it can be valuable information. Using gained data to find fraudulent activities will help government to get more tax money from people that are not paying the right tax. It can really help out the economy. Using gathered data in a grocery store and processing it to become some good information is something each company all need to strive to do. They need to analyze data in the database to improve the need of customers.

  3. Good article. And yes, data mining is a pretty crucial part of any business. I found the diaper and alcohol example pretty interesting.

Comments are closed.