4th amst-R-dam meetup

Hi folks,

We have two more excellent talks lined up for our next meetup. Details follow:

First talk will be on The Semantic Web, R, and Enipedia.tudelft.nl (http://enipedia.tudelft.nl/) by Chris Davis and Alfredas Chmieliauskas, TU Delft

Abstract: Depending on who you ask, the Semantic Web is either going to change the world, or is an over-hyped techno-utopian vision. Based on our experience, we suspect that the former interpretation is starting to become reality.

This talk will cover the basics of the Semantic Web in terms of its tools and philosophies. We will give general examples using DBpedia, a project based on structured data extracted from Wikipedia, and show how the SPARQL library for R can be used to work with this data.

From there, we will discuss applications in our own domain of energy and industry, and will show how on enipedia.tudelft.nl (http://enipedia.tudelft.nl/) we are taking existing public data sets, converting them, and linking them together to allow for much richer and more efficient analyses to be done.

Chris will talk about how he is integrating R for his work on semantic web. More details about his talk will follow shortly.

Next, Richard Gill from Uni. of Leiden will talk on two subjects.

Richard's first talk: R on the iPhone (or iPad). Yes, it can be done. After all, this is a pocket computer which runs on BSD Unix, with the capacity of a decent desktop computer of 10 years ago. First we must jailbreak (or “root”) the machine, which is an adventure in itself. After that we just have to install R … fortunately the hard work has been done by others (thanks especially to “ech0chrome”).

Richard's second talk: reverse engineering the logo of the Dutch statistical society. Some thirty years ago, artist Tobias Baanders (house graphical designer at CWI - Centre for Mathematics and Computer Science, Amsterdam) created a new logo for the society - an artist’s impression of statistics and OR. It was drawn free-hand in pen and ink. Later, a scanned image was converted to a rather imperfect postscript file containing a long list of jerky spline curves. I used to give my students the assignment “fit a family of smooth curves, preferably with a statistical story behind them, resulting in a compact description and scalable vector graphic file”. The challenge was also offered open to all (with a prize) on my home page. After several years of no progress, I turned to the problem myself. My solution involves R’s three dimensional graphics package rgl. I hope to show you a movie of a living 3d logo.

Looking forward to continuing our momentum of wonderful talks and seeing you folks at the meeting.

Cheers, Rahul + Werner