R community well known, R-bloggers founder, Tal Galili, will show you a few programming skills from Creating beautiful trees of clusterings to updating R.
This event is co-organized with NYC open date meetup group http://www.meetup.com/NYC-Open-Data/ and both group will handle registration. If you are members of both groups, make sure you only sign up once. Thank you!
Squarespace will generously sponsor us their office space for this event. The event is held at 459 Broadway(across Grand Street), New York, NY 10013. It is on Floor 5.There's an elevator entrance around the corner on Grand street. Special Thanks go to Joel Witten and his colleagues!
Mr. Tal Galili is an applied statistician and an enthusiastic R user. He holds a masters degree in biostatistics and is a lecturer for statistics courses at Tel Aviv University, while working on his PhD in statistics there. Tal writes the blog R-Statistics.com, and is the founder of R-bloggers.com: a central hub (aggregator) of content collected from bloggers who write about R.
If you wish to play with the code a bit during the talk, you are invited to come with your laptops - have R installed (if you don't have an IDE yet - RStudio is great), and run the following sequence prior to the talk:
if (!require('installr')) install.packages('installr'); require('installr')## install.Rtools() # run this if you are using Windows and don't have Rtools require2(devtools)install_github('dendextend', 'talgalili')require2(Rcpp)install_github('dendextendRcpp', 'talgalili')
The code of the dendextend talk will here, for you to follow along:https://github.com/talgalili/dendextend/blob/master/inst/doc/2013-09-05_Boston-useR/2013-09-05_Boston-useR.R
7:00-8:30pm Tal Galili will give the talk
8:30-9:00pm Q & A
There will be a combination of four talks:
1) A biostatistical talk about "deciphering the tRNA code" - an interesting case study of using R for reaching an interesting discovery about the way the basic tRNA molecule is taking part in the centeral dogma of molecular biology. This talk will involve a combination of CART, cross-validation, heirarchical clusterings - and many figures. (this talk was recently given in useR 2013 and before then in useR 2010)
2) Introducing the "installr" package - for updating R, installing software with it, analyzing log files of R packages downloads, and some other minor tasks in R. This is an extended version of the talk recently given in useR 2013.
3) The recent work on the package "dendextend" (and dendextendRcpp) - intended for visualizing and comparing trees of heirarchical clusterings (a.k.a: dendrograms) with R.
There will be a tutorial which will go through a short overview of heirarchical clustering, "hclust" and "dendrogram" objects in R and then introduce the new abilities of manipulating trees using the "dendextend" package. From this talk you will know how to create, change, visualize, and statistically compare two trees of heirarchical clusterings (with some sprinkles of Rcpp).
4) (As time permits), some words about blogging about R. (Yes, it is about my favorite website: http://www.r-bloggers.com )