2
$\begingroup$

I am COMPLETELY new to the field of Data Science, mainly because every employer I have worked for, simply COULDN'T sell any customers anything that would use techniques learned in this field.

Of particular interest to me is machine learning/Predictive Analysis.

I have attempted many "test projects" myself, but I seem to NEED some sort of outside "catalyst" to tell me a specific goal, and a specific set of guidelines, when I am trying to learn something. Otherwise, I tend to lose focus, and jump from one interesting topic to the next, without ever gaining any experience.

Thank you!!

$\endgroup$
  • $\begingroup$ I've answered a similar question a while ago. You can check my answer here: datascience.stackexchange.com/a/843/2452. $\endgroup$ – Aleksandr Blekh Jan 22 '15 at 11:42
  • $\begingroup$ This is too broad to be an effective question here. Maybe you can list a couple projects you are interested in, what you've tried, and name the obstacle you face in pursuing each. $\endgroup$ – Sean Owen Jan 23 '15 at 10:23
3
$\begingroup$

I would suggest Kaggle learning projects - http://www.kaggle.com/competitions

Look for the ones in the 101 section that offer knowledge. There's many pre-made solutions ready, which you can ingest and try variations of.

Also, I have bookmarked a Comprehensive learning path – Data Science in Python, which among other things gives a few answers to your specific question.

$\endgroup$
  • $\begingroup$ Thank you for your specific response to a rather vague question. This was EXACTLY what I was looking for! (I don't have enough Rep to "Vote Up" $\endgroup$ – Steve Rowland Jan 23 '15 at 15:13
2
$\begingroup$

You can use some public data sets to play around with.

The iris flower data set is great for classification problems. As an example you can try to classify the flower species with k-nearest neighbor model or a decision tree (CART) model.

You can find this and other sample machine learning datasets here: http://archive.ics.uci.edu/ml/datasets.html

This list identifies what type of machine learning tasks can be performed with each dataset.

$\endgroup$
  • $\begingroup$ Thank you very much. I would "Vote Up" this response, but I do not have enough Rep. $\endgroup$ – Steve Rowland Jan 23 '15 at 15:14

Not the answer you're looking for? Browse other questions tagged or ask your own question.