Edit
Tony Hirst on Data Wrangling
Tweets and tips inspired by the #dalmooc guest hangout on Monday October 27, 2014
- Reference to John W. Tukey, "We need Both Exploratory and Confirmatory", The American Statistician, Vol 34 No. 1 (Feb. 1980), pp 23-25.
Tukey: “The only way humans can do BETTER than computers is to take a chance of doing WORSE.” http://bit.ly/1xsMnrQ- On iPython Notebooks - which was "before topic" but relevant to what followed
- @martinstabe Notebook cells can run code from variety of languages (python, R, javascript, etc) eg R http://bit.ly/1xsRQPp
- @martinstabe nbviewer gives html preview; actual notebooks are interactive, cell based docs; code cells can be executed one cell at a time
- The Hangout
Today, 3 pm Central (dallas) time: Tony Hirst (@psychemedia) will present on messing with data #dalmooc https://plus.google.com/u/0/events/c50ojgl3bsl8fa8kll5puclmbvk …
Waiting for @psychemedia on "Data Wrangling" to start on Google Hangouts https://plus.google.com/u/0/events/c50ojgl3bsl8fa8kll5puclmbvk … (#dalmooc)- "we shouldn't be afraid to put powerful tools in the hands of individuals who might not quite understand them yet" @psychemedia #dalmooc
- Tools - openrefine.org
Listening to @psychemedia talk about cleaning data as part of #dalmooc cc: @edXOnline @gsiemens #mooc
Listening to the ever excellent @psychemedia explaining data wrangling for #dalmooc: https://plus.google.com/u/0/events/c50ojgl3bsl8fa8kll5puclmbvk …
@psychemedia on Google Hangout on data wrangling - https://www.youtube.com/watch?v=D6t4eztDveU#t=1115 … #openrefine.org #dalmooc via @itsmaloy- Sankey diagrams at #dalmooc google hangout presented by @psychemedia Random examples: https://www.google.com/search?q=sankey+diagrams&rlz=1C1SAVK_enUS531US531&espv=2&biw=1353&bih=732&tbm=isch&tbo=u&source=univ&sa=X&ei=Z65OVImCFMyIsQTdp4GIDQ&ved=0CCYQsAQ …
- Nice example presented by @psychemedia http://www.coolinfographics.com/blog/2014/8/29/false-visualizations-sizing-circles-in-infographics.html … False Visualizations: Sizing Circles in Infographics #dalmooc
- For folks interesting in looking at @psychemedia's posts, recipes, see here: http://blog.ouseful.info/ #dalmooc
RT @eRomanMe: @psychemedia on Google Hangout on data wrangling - http://youtube.com/watch?v=D6t4ez … #openrefine.org #dalmooc via @itsmaloy
@psychemedia The video dies sometimes; maybe sharing the slides is useful: http://docs.google.com/presentation/d/1-hwAoo6dF7FlT7ZSzEcqZFiu25F0pLgdMqhb8mD3_yU/ … #dalmooc- and they were shared ... see 21:34
- Tools - gephi.github.io
- Tools R and gplot
- R (+gplot) statistical and graphical programming language: http://www.statmethods.net/index.html presented by @psychemedia at #dalmooc
- Tools iPython notebook
- Excellent Google Hangout on Data Wrangling by @psychemedia should be available on YouTube very soon at https://www.youtube.com/watch?v=D6t4eztDveU …. #dalmooc
- #dalmooc Apologies for all the F1 data - I’ve been submerged in http://bit.ly/1eUPu2A for the last week and not been up for air!
- #dalmooc An example of one of my conversations with data from earlier this year… http://bit.ly/1rtcE8D
- #dalmooc another example of a data conversation… http://bit.ly/1puSDhA
Interesting presentation by @psychemedia despite some glitches. A good overview into the complexities of data representation. #dalmooc
@psychemedia Tony Hirst: more inspiration to learn R and Python. Thanks. #dalmooc- And post hangout additions from the man himself....
- Tools: R, python, pandas, RStudio and iPython notebook
- @jimacunningham R and python/pandas have a similar notion of dataframes; RStudio and IPython notebooks have v powerful workflows #dalmooc
- #dalmooc My slides here - http://slidesha.re/1DTZBC6 Didn’t have time to annotate them unfortunately:-(
- Tutorial on on Gephi
- #dalmooc Gephi tutorial available here: http://bit.ly/azedET
- ... and and example of openrefine
- And a series of great iPython notebooks ...
- #dalmooc writing interactive pivot tables into IPython notebooks… http://bit.ly/10JwZNV
- Tools: data2txt
- #dalmooc And finally, s/thing I didn’t talk about but that will get more widespread…? data2txt - v simple example: http://bit.ly/1ry7HMW
- With reactions from folks who watched the recording
Listening to @psychemedia Hangout from #dalmooc on data wrangling. Learning the ropes rather than flying by the seat of my pants.- @rockenschtroodl Many Eyes at http://ibm.co/1rwP1IK This pivot table widget http://bit.ly/1rC7BDW can generate treemaps too
