Do not be run by data in Ai-Dan Rose he

Being directed by the data is usually used and understood with positive connotations, but when I hear the word used I am a little concerned about the “directed” decisions that may occur. Let me explain why.

According to the data of Wikipedia driven by the data means “the adjective directed by the data means that progress in an activity is bound by the data, rather than by intuition or personal experience.” In other words – see data as a major source of information to act. When the data give you a reason to act, you act. At a glance it can seem like a very healthy way to work and especially in the field of it, in many ways to rely on data. But in fact being directed by the data can be very problematic when working with him. I actually think people who say they are directed from data are generally on the wrong path. This does not mean that I am against trying to understand your data. I am actually a great believer that collecting, understanding and preparing data on projects should be activities with the largest resources allocated to him. So I’m in favor of good data science, but against data direction and see it as two very different things.

But then why is it so problematic to be driven by data?

My main argument is that the driver standing after decision -making and activities should not be the data you have, but on the contrary, curiosity about the problem and the world around him. In a sense that would mean to be directed to the data you do not have. The ultimate goal of he often is to solve a problem or improve a process and solutions for them do not always exist in the data you have generated or are being generated by current solutions of the world. So, instead, you have to be directed by curiosity or at least directed by the problem. This means that you should not approach the problems by seeing your data and making a conclusion. You need to look at your data and look for the blind points and from there to be curious. What is what you don’t know? I will return to curiosity later. First I have some other arguments against being directed by the data.

You will rarely have all the important data for a problem. Even after the exhaustion of all possible data sources. So when you make conclusions from the data you have, the finish will at least always be a little far away. This is not to say that the data is not useful and that the finish is not useful, but you will always be at least a little wrong. As the statistics would say, “All models are wrong, but some are useful.”

Another problem with being data -driven is that there is a story that the decisions made in the data are better than the decisions made on the sense of intestines. And while this can be true sometimes, the data is not one -way and can be very useful sometimes and very deceptive for others.

An example is the father of modern statistics Ronald Fischer who was also more or less directed at Hindsight. He stubbornly concluded that the data showed that lung cancer was not the result of smoking. The correlation he said should be vice versa and people with lung cancer or higher risk of lung cancer are simply more likely to be smokers. He argued that it was either a genetic connection or that patients with cancer would use smoking to relieve lung pain. Even the best statistics can be told stories that are far from the truth from the data.

The last problem with the data is his ability to tell you the story you want to tell. This can be done consciously or unconsciously. A famous quote by the Ronald Case Economist Ronald goes “If you torture the data long enough, it will confess nothing”, so there is no certainty that the conclusion you get from the data is correct. Interpretation can be very one -sided and sometimes we torture data even without being aware of ourselves.

For curiosity

So as it was promised I was returning to be curious. If I had to choose a keyword to succeed with it it would be curiosity. He usually starts with a process to optimize or a problem to solve and before training a model on the data you need to be curious about the problem. In that way the data then come to the problem and as a result will be more important and more specific to the problem.

Curiosity for me means exploring as few prejudices as possible. The best example for me is when children raise rocks on the ground just to see what is under rock. If you have ever seen a child doing what you will have seen that there is no expectation, only excitement both before and after the rock and is raised. And this is exactly what makes the practitioners’ curiosity. It leads to excitement which in turn leads to passion. Passion makes everything much easier and even tedious parts of a project will feel easy.

He is also explorer in its nature and that is why it suits it so well that it is curious. If there is specific expectations in an exploratory process, then it is almost disappointed.

As a result, you should allow curiosity to be the main driver behind the decisions and activities you make. Being directed by the data is reactive in nature and if you want to be innovative in problem solving you need to be proactive. Being proactive requires you to be curious about your blind and prompted by the unknown.