It's actually incredibly hard, especially if you want to achieve better results than with a current 'gold standard' technique/algorithm, applied on your particu

It's actually incredibly hard, especially if you want to achieve better results than with a current 'gold standard' technique/algorithm, applied on your particular problem. While the article doesn't have this title (why would you even choose one with such a high bias?), I presume the submitter decided upon this title after being encouraged by this affirmation of the article's author: 'This data indicates that the skills necessary to be a data “wizard” can be learned in disciplines other than computer sciences and mathematics.'. This is a half-baked conclusion. I'd reason most Kaggle participants are first of all, machine learning fans, either professionals or 'amateurs' with no formal qualifications, having studied it as a hobby. I doubt people with a degree in cognitive sciences or otherwise in the 'other' categories as mentioned in the article learned enough just through their university studies to readily be able to jump into machine learning.