tgkolda / genetic_scheduling Goto Github PK

Hi @amklinv-nnl -- out of curiosity, I took a look at the citations.yaml file. I was interested in the ones with high citation counts because it strained credibility that someone could have 500k citations -- alas, that's actually real! :-)

So I looked a bit down the list. The ones with 100k+ citations are these:

100577 Mark Adams
111531 John Bell
111847 Klaus Roppert
123618 Hong Zhang
125749 Yang Liu
135462 Jun Wang
141758 Jeffrey Sachs
144669 Maximilian Ruth
179395 Zhen Zhang
190442 Ioannis Papadopoulos
457895 John Alexander
500147 Ryan Richard

There are some I can't find on google scholar or for which I suspect that the numbers aren't quite real:

• I believe that Ruth is actually a grad student (https://www.cam.cornell.edu/research/grad-students/maximilian-ruth)
• I believe that Roppert is a postdoc (https://scholar.google.com/citations?user=c39ym2kAAAAJ&hl=en&oi=ao)
• John Bell (according to the minisymposium spreadsheet) is at LBL, but not the same as the highly cited one here: https://scholar.google.com/citations?user=YgKvUVcAAAAJ&hl=en&oi=ao
• Hong Zhang is at Argonne (https://www.anl.gov/profile/hong-zhang) and his website links to this google scholar profile: https://scholar.google.com/citations?user=lo_niigAAAAJ&hl=en It shows 5174, not 123k citations.
• There is a Ryan Richard who really is that highly cited, but that's not the speaker I believe. The highly cited one is in psychology, whereas the minisymposium speaker is talking about molecular dynamics -- possibly the same person, but probably not. I suspect it's actually this one: https://scholar.google.com/citations?user=QonezmQAAAAJ&hl=en with 5400 instead of 500k citations.

I really don't think that it matters very much for the end result whether these numbers are correct or not. I'm mostly curious how you scraped them.

As a separate issue, if you assign room sizes based on cumulative citation counts of speakers, then the few very very highly cited people will trump all others. If you just truncate every minisymposium's accumulated citation count at, say, 20k, then you eliminate some of the distortions by identifying the wrong person as a deciding factor.

I was trying to see whether we can produce "marked-up" .md files containing the schedule to improve readability, but I find that I don't actually know how to run the software. If you give me a few lines of instructions, then I will convert this into text for the readme file in this directory.

I suspect that your goal is to eventually hand this project off to the people running the next conference. It might be useful to them to have documentation on what to put where and how to run things. I'm happy to convert informal hints into formal documentation.

One way of estimating session attendance is to take a poll among the participants. Here is a description of using the outcomes of a poll to arrive at a attendance-maximizing schedule used for scheduling some scientific conferences, regards, Frits Spieksma

Vangerven, B., A. Ficker, D. Goossens, W. Passchyn, F.C.R. Spieksma, and G. Woeginger, Conference scheduling – A personalized approach, OMEGA 81, 38-47.