3

If I have a set of data in whatever format, CSV, JSON, etc.

I wanted to throw it at machine learning algorithms and be able to extract some correlation facts from these datasets.

Some of the things that the software might tell me:

  • On average bugs are reported 20% more on Tuesday.
  • There was a spike in bug reports in December.
  • More bugs were reported when we had more outages

Or anything that is useful about the data I pass.

What software can be used for that type of information gathering from data?

  • So, the structure of the data is unknown, and you don't want to inspect the structure by yourself, you want the program to discover it and understand it by itself, is my understanding of your question correct? – Nicolas Raoul Jan 18 '16 at 7:52
1

This is the sort of think that the python data exploration and visualization tool chain Pandas is excellent for - you will have to do a little exploring of the data yourself but you will be amazed at what you can find out quite rapidly.

If you would like to go the whole hog of machine learning then there is the SciKit Learn Machine Learning in Python package.

If you are running windows and would like a single download to install the lot then you could try Python X-Y.

  • You don't mention your platform but that is OK all of the above is available for just about any platform.
  • Budget - All of the above is free, gratis & Open Source
  • Lots of books out there on using the above.
  • Lots of community support and Stack Overflow help on the above as well. Currently I get 520,336 questions tagged [python], 17,023 questions tagged [python][pandas] and 3,820 questions tagged [scikit-learn] on Stack Overflow
| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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