The balancing act of data mining ethics: the challenges of ethical data mining

When most people think of data mining, one of the first things that comes to mind is the scandals surrounding data privacy. Consumers are becoming increasingly aware of how their data is being passed around behind the scenes. The lack of data mining ethics in larger organisations has become a contentious issue.

The controversy over data mining really picked up steam when it was discovered that Cambridge Analytica had paid Facebook for access to the data of over 50 million US Citizens in preparation for the 2016 US election. Since then, Facebook has drawn consistent criticism for its data mining practices. Criticism has ranged from disillusioned users in the #DeleteFacebook campaign to commentators suggesting that the company lacks an “ethical roadmap.”

In the aftermath of the controversy over the 2016 election and the questionable data collections practices across Silicon Valley, data mining has become a dirty word. Today, data mining has become synonymous with selling off user privacy for financial gain. Even after the GDPR, data mining practices lag behind consumer expectations.

Just recently, Epic Games, the gaming juggernaut behind Fortnite came under fire from users in a Reddit thread that claimed the company was mining user data from the Steam accounts of users. Dan Vogel, Epic Games’ Vice President of Engineering attempted to address the criticism saying that “we only import your steam friends with your explicit permission”.

Vogel’s response did not satisfy the player base who replied and asserted that no such permission had been given. Epic Games found out first hand that even though you have legal permission to access user data, you may not have the active approval of your users. Epic Games are now one of many companies who find themselves on the wrong side of a dispute over data mining ethics.

Other enterprises on the sideline are left scratching their heads as they attempt to define how to access user data for mining in a way that is ethical, both in a legal sense and contextually to users. Today, the legal right to data is a thin shield against the court of opinion. If users feel that a company has accessed data unethically, they are going to speak out even if the data was obtained legally.

Fighting digital fraud with the power of data mining

Dr Stephen Topliss, VP, Fraud and Identity at LexisNexis Risk Solutions, discusses the power of data mining in combatting fraud

The challenges of ethical data mining

The central challenge that underlines ethical data mining is that there is no collective standard of data mining ethics to point to. For instance, the United States has no legal definition of personal data. While the regulations established with the GDPR did launch somewhat of a legal framework in Europe, having legal consent for data collection is not enough to protect against alienating users with “unethical” data usage.

Currently, the line in the sand between ethical and unethical data mining is being drawn by allowing users to opt-in or opt-out of data collection. As Doug Welch, Baylor University’s chief privacy officer states the data ethics debate “the primary [data ethics] debate is whether the choice is made through an ‘opt-in’ or ‘opt-out’ decision.”

Welch continues that “opt-in requires an active indication of consent; opt-out automatically makes the user a part of the system with an opportunity to remove themselves.” Under these systems, the user has control in whether they partake in a service or not.

However, opt-in systems represent certain challenges to modern enterprises. For instance, “from a business standpoint, though, the number of people who participate under an opt-in choice is generally lower because the notice often scares potential users away…but that’s somewhat the point of making an informed choice.”

Companies therefore have to make the choice between being transparent and warning users about data mining practices. However, even if they notify users, they need to go the extra mile to highlight their data collection activities. Organisations that fail to do so will face harsh criticism and leave themselves vulnerable to accusations of misleading users.

How to use process data mining to improve DevOps

If the value of efficient business processes is obvious, it makes sense to us process data mining to improve DevOps. Here’s how you can…

Data mining ethics: the responsibility of private organisations

In the field of data mining, legal data collection is no longer enough to placate public opinion. Data collection practices must also be perceived as ethical and transparent as well. While broadcasting data mining practices with large opt-in notifications isn’t appealing to the bottom line, alienating customers by obscuring data collection practices isn’t either.

Any company that engages in data mining, should seek it has not only the legal right to access data but the explicit permission of the user. The former is a legal necessity, the latter is essential for meeting consumer expectations. Making sure that customers are aware of data collection is essential for reducing the risk of friction.

An increase in transparency and communication with customers is the only way to navigate the current climate’s lack of legal clarity. While organisations can do the legal minimum, public opinion will be unforgiving if a company’s data collection practices are thought to be unethical.

Though data mining ethics are improving, many companies are souring relationships with customers with poor data collection practices. For now, it is the responsibility of private organisations to protect themselves from scrutiny by becoming staunch data ethicists.

 For more on ethical AI

Related Topics

Data Mining
Ethical AI