[Pycon] [new paper] "Pietro Battiston" - Data analysis in Python - come in, don't get lost

[info a pycon.it]

Title: Data analysis in Python - come in, don't get lost
Duration: 45 (includes Q&A)
Q&A Session: 15
Language: en
Type: Talk

Abstract: "_I think the problem, to be quite honest with you, is that you've never actually known what the question is_", says Deep Thought in The Hitchhiker's Guide to the Galaxy.

The Python ecosystem for data analysis has flourished in the last years, but as a consequence, choosing the right tool for the job is not always simple.

This talk will briefly draw a landscape of some of the most relevant libraries, and of their different aims - getting a glimpse of the daily workflow of a "pyconomist".

**What to expect**

 - a _really_ introductory level, hands-on, journey through what data analysis does... in Python
 - a comprehensive view of the "Pydata stack", that is, Python libraries allowing us to analyze data (with specific attention to _social_ data)
 - an emphasis on the kind of _question_ that each software "asks" your data, like contrasting explanation vs. prediction (and on the fact that _it makes sense_ to ask multiple questions)
 - hands-on simple examples showcasing the approach of different libraries

**What _not_ to expect**

- a deep analysis of any of the libraries
- a thorough technical investigation of any specific theoretical problem, or application

Tags: [u'numpy', u'data-analysis', u'matplotlib', u'pandas', u'data', u'sklearn', u'pydata']