Any members of kaggle.com?

Question

data mining

The experience is interesting:
- is it easy to start
- are commands needed
- what algorithms are usually used (if this is not know-how :) )
Thank you.

Answer 1

“It's really hard to get started. It takes a very long time to enter the task.
- In any case, you need it, if you are not a genius. It is very easy to get stuck on some algorithm. You need a strong mathematician + coder at least.
— The algorithm is highly dependent on the task. But, in general, in addition to the standard set of datamining, it would not be bad to know computational mathematics (gradient methods, boundary value problem, transopter problem, etc.) and standard methods for solving standard problems.

Answer 2

I agree about the team. But "their own" and nearby, it's a luxury :). I think I'll try on a simple contest, and then how it goes.
As for semantics, I don't know. I'm more on the side of statistics, it's easier. I honestly tried to get through homepages.inf.ed.ac.uk/jbos/comsem/ , but did not understand how to induce something formal from the text, like first-order logic. For small texts and grammars, it passes, but it doesn’t get beyond toy examples. So statistics and ML look more productive.

Answer 3

So I began to write a detailed answer to the question “Is it easy to start?”
habrahabr.ru/post/165001/