Big Data has been a huge revolution in many fields such as business management, healthcare and marketing. Decision-making automation is designed to delete the human factor from the equation. This way, we make sure that we get rid of intuition and act based only on objectivity.
When we do this, we pass all the responsibility on to mathematical algorithms. They search through huge mountains of data and make decisions that affect our lives, but we have no clue about why or how they’ve done so, and this may be really disturbing. It is a highly complex system, and even those who embrace it eagerly have their qualms.
Steve Lohr explains it perfectly in his article ‘If algorithms know all, how much should humans help?’. He wonders whether human supervision is needed to ensure that the data are making the right decisions.
Lohr raises a very important question, ‘Will a technology that promises large benefits on average sufficiently protect the individual from a mysterious and wayward decision that might have a lasting effect on a person’s life?’ This may not be a challenge when it comes to marketing, but things change in other fields like healthcare, where it is vital to make the right decisions.
A more human Big Data
Regarding human intervention in decision-making processes with Big Data, there are two main different positions on the matter:
On the one hand, some people think that the storytelling is the key. We’re not talking about the narrative, but some kind of understandable explanations of how those automated decisions were made, how they relate to us and what part of such decisions came from the human and which one came from the machine. In short, a human interpretation that will let us know for sure that we’re doing things right.
On the other hand, others may think that the human factor only stops the algorithmic systems from working. They support the idea that introducing the human bias will prevent decisions from being automated so they will continue to be what they have been so far: data analysis outcomes.
In Lohr’s opinion, the best solution is halfway between those two options. The algorithms’ creators should be able to refine them. Not to get the maximum efficiency and benefit from them, but to give the greatest responsibility to the individual and reduce the risk of making mistakes. The objective is not to see it from a human point of view, but to improve the quality of the results offered to people. The human factor may provide the subtle nuances that the Big Data technology cannot see. Then they could work together much better than how they would do on their own.
Translation by Susana Castro
Image: Jared Tarbell