A question for my IR friends -- does anyone know of a curated list of datasets for IR? I'm specifically interested in open user study data sets (sets of tasks, queries, and performance).
Henry Feild retweeted
Endicott's @girlswhocode club is looking for students! Girls 6-12th grade, learn to program from 4-5pm weds.; see sites.google.com/endicott.ed…
Anyone know how mouse clicks are recorded in Tobii Studio 3.4.5? mousedown+mouseup? Is the timestamp and (x,y) from the former or latter?
This is pretty cool (minus the privacy implications) techcrunch.com/2016/11/23/se…
Can anyone point me to studies/surveys reporting the number of people worldwide who are proficient in programming (or a surrogate stat)?
Henry Feild retweeted
One of those notable moments where we have an IR researcher doing Science AMA. xitter.cf/elad1/status/772792551…
I'm doing an "Ask Me Anything" on Reddit today: reddit.com/r/science/comment…
Very interesting in scope/scale even if findings have been said before. I'm sure CS is in a similar boat. xitter.cf/vardi/status/772432820…
Why Do So Many Women Who Study Engineering Leave the Field?
hbr.org/2016/08/why-do-so-ma…
Henry Feild retweeted
Happy to announce @diegoceccarelli from Bloomberg will be giving a presentation on Learning to Rank at #Lucene4IR bit.ly/lucene4ir-site
Henry Feild retweeted
Want to come to the #Lucene4IR workshop in Glasgow? Travel grants are available, visit bit.ly/lucene4ir-site for details. – at Glasgow, Scotland
Henry Feild retweeted
Happy to announce that Charlie Hull from .@FlaxSearch will be giving a keynote at the #Lucene4IR workshop bit.ly/lucene4ir-site – at Glasgow, Scotland
Henry Feild retweeted
The #Lucene4IR workshop website is now up at bit.ly/lucene4ir-site We have 20 free registrations & travel grants thanx to #ELIAS – at Glasgow, Scotland
Henry Feild retweeted
It's ML, not magic: simple questions you should ask that will help reduce AI hype
(minimal ML/AI knowledge required)
smerity.com/articles/2016/ml… – at California, USA
And this showed up in my feed right after my last post, and it's too related to pass by :) xitter.cf/arstechnica/status/750…
Science funder becomes publisher; publishing, reading articles both free arstechnica.com/science/2016… by @j_timmer
Intersting article on research of peer review nature.com/news/let-s-make-p…
Henry Feild retweeted
Are you working with Lucene/Solr for research or teaching? If so we are running an #elias workshop #lucene4ir in Glasgow in September.
Henry Feild retweeted
“The alleged problems with SIGIR” by Fernando Diaz medium.com/@diazf/the-allege…
Henry Feild retweeted
My student is using genetic algs to generate new recipes. Today he baked apple cake based on an outputted recipe
Henry Feild retweeted
Voting on my pocket-tweet continues. There's still time to cast your vote or to explain to me why you did so
Henry Feild retweeted
Dr. Lawrie presented on 2006 ICPC Most Influential Paper (w/Binkley, Morrell, Feild) on identifier styles in code: felienne.com/archives/4976
Henry Feild retweeted
Congratulations Drs. Lawrie, Binkley, Morrell, & alum @hankfeild for their 2006 ICPC paper being chosen as ICPC'16 Most Influential Paper!












