Data Mining - A Blessing or a Curse?

Data mining (not to be confused with the super popular Bitcoin mining) has rose from obscurity in recent years. As is the case with artificial intelligence and machine learning, increased computing power has allowed data mining to blossom in this decade. So, what exactly is data mining?

Well, simply put data mining is extrapolating trends, patterns, abnormalities, correlations and more in order to predict future outcomes. As a result, data mining requires large amounts of data, and a computing system powerful enough to process it. You already probably see the overlap of data mining and cloud computing, big data, statistics, and artificial intelligence. The innovations of these fields have allowed data mining to become what it is today.

However, data mining has received a negative connotation. This is because marketing, retail and a wide variety of companies have used data mining to predict what you will search or buy, and it is often scarily accurate. For example, Target used the data they acquired from retail shoppers and created an algorithm that predicted when a certain customer was pregnant. Target would then send coupons containing items a new mother would require. However, in order to hide themselves, they sprinkled random coupons in the list as well. After data mining and implementing this algorithm, Target had several cases where they predicted pregnancies and sent coupons to people before they knew themselves. There is a very interesting case where a father received coupons for newborn items targetted to his daughter, who was pregnant, unbeknownst to him. You can read more about it here. As you can see, data mining is very powerful with large datasets, and can be useful if used correctly.

However, there are two sides to the coin known as data mining. It can be used for good, such as fraud detection. Do you ever wonder how banks know when to call you if a suspicious purchase is made? Well, you can thank data mining. Banks use data mining to determine whether or not a purchase was likely made by you. Although every person is different, each person has their own purchases, and in the bigger picture, they tend to have a pattern. Banks use data mining to explore these patterns, and single out a discrepancy. This analysis happens in seconds, where a card can be declined because the purchase is suspicious. As a result, banks have saved customers millions, if not billions of dollars due to the efficiency and accuracy of the fraud algorithms.

Data mining, like most things, has benefits and drawbacks. Because it is so powerful with large datasets, companies can misuse it to invade people's privacy, such as the case with Target. However, it can also be used to protect people or further positive initiatives, as with the case of fraud. Consequently, the government and people must decide how to regulate data mining, and the issues that can be solved using it. Data mining is still expanding, and the future's privacy depends on the regulation of it. All in all, I have a positive outlook on data mining, I believe that with the right tools and datasets, we can use data mining to revolutionize the world and its efficiency.

Inspiration + Read More