Add new tensor consisting of counts of words in abstracts published on arxiv.org

[etphipp]

This adds the meta-data for a new count tensor consisting of counts of words appearing in abstracts published on arxiv.org, and displays some interesting patterns in CP analysis. I don't currently have a publicly accessible URL for obtaining the tensor and mode mappings, but can provide an a URL by email to whomever will move the files to the FROSTT server.

[jeewhanchoi]

Hi Eric,

Sorry I didn't realize there was a pending pull request. Shaden owns (and pays for) the server for hosting existing files on FROSTT and I am trying to figure out if we can specify different servers for storage.

I'm going to test out if I can use a Dropbox link for some new tensors I want to add to FROSTT. Once I get that working, and if you are still interested in adding the arxiv tensor to FROSTT, I'll merge and ask you for the files.

Thanks.

-- Jee

[etphipp]

No worries, I got the impression nobody was monitoring it. I would still like to add the arxiv tensor because multiple people have asked for it.

[jeewhanchoi]

Okay, I modified the website so that we can use any arbitrary URL to link the files.
Send me an email with the link to the files and I'll add them to my Amazon AWS site and host it from there.
Thanks!

[etphipp]

Sounds good. We have a paper that should be submitted in the next month or two that uses this data set, and my understanding is Sandia will host data sets that are associated with a publication. I'll let you know when that happens.