Dominika Tkaczyk

Tired of the academic grind, but still want to make a difference? Join @CrossrefOrg as a Principal R&D Developer where there is no REF and you don’t need to worry about your IF, but you’ll still make a big impact building Open Scholarly Infrastructure! crossref.org/jobs/

Come help @CrossrefOrg design and build a new generation of open scholarly infrastructure. We are hiring a Principal R&D Developer. crossref.org/jobs/

.@SDProc will be online following @emnlp2020's announcement today. Workshop date moved to *Nov 19*. Updates to changes to #sdp's submission deadlines will be announced. #clscisumm #laysumm #longsumm shared task registrations bit.ly/2JbTFP8 to remain open till April 30

Proposing 1st workshop on Scholarly Document Processing SDP 2020 @ @aclmeeting. Calling #nlproc #ir #ml #scidoc community to support and vote for us! With @robodasha @anitawaard @ehovy @Philipp_Mayr @petrknoth @TirthankarSlg @d_tkaczyk @RobertMPatton Dayne Freitag #birndl #wosp

@aarontay @CrossrefOrg Yes. "doi-asserted-by" tells you who provided "DOI" field for this reference: either the publisher (and we don't need to use any matching algorithm), or Crossref (using Crossref's matching algorithm).

@d_tkaczyk @CrossrefOrg ic. so in cases where "doi-asserted-by" = publisher? There is already a doi field given by publisher?

You are probably not Garry Kasparov’s child, but if you’re into #citation #metadata, you may be keen to see @d_tkaczyk’s latest evaluation `Reference Matching: For real this time` blogged at crossref.org/blog/reference…. And check out the open analyses data at github.com/CrossRef/refer…

@aarontay @CrossrefOrg "unstructured" field is just the input for the matching, the string representation of the reference. When there is "doi-asserted-by" = "crossref", ref's "DOI" field contains the target DOI found by our legacy matcher.

@CrossrefOrg @d_tkaczyk Helpful. I read a couple of posts up chain. I just happen to be playing with the crossref apis to look at references. So if I see a "unstructured" ref it is something crossref can't find a doi to link to? "doi-asserted-by":"crossref" are those found by the proccess in blog?

@keswconference Are you still accepting poster submissions? It's not clear from the website.

@paul_nu Hi Paul, please email me at d.tkaczyk(at)icm.edu.pl You can also report an issue on Cermine's GitHub.

My #Jurix2015 poster available online dx.doi.org/10.6084/m9.fig…

@HuaweiUSA One of the largest #Spark clusters in EU + a seasoned #DataScience team + industry partners w/ #BigData. Stay tuned for results!

Structured affiliations extraction in #CERMINE — an article in the latest D-Lib Magazine dlib.org/dlib/november1…

Random Ferns implementation in @ApacheSpark + a relation from #SparkSummitEU. A blog post by @pjden #MachineLearning #BigData #OpenSource

Our GROTOAP dataset (Ground Truth for Sci Pubs) has a DOI. It's the 1st dataset in @OpenSciPlatform's RepOD dx.doi.org/10.18150/96480… #opendata

@Francopoulo Currently no, only the text parts are processed. We'd like to add this functionality some day in the future.

Two of us leveled up this month: @kobosm is a PhD (see link), @mjungiewicz — a MSc. Congrats! adalab.icm.edu.pl/blog/news/2015…

@petrknoth Hi Petr! A minor difference: CERMINE extracts journal/volume/issue/pages while AFAIK GROBID doesn't #WOSP2015

Impressions from #ESWC2015 by @bolikowski (incl. bragging about the award for CERMINE) adalab.icm.edu.pl/blog/news/2015…

Today at 2:30pm our #machinelearning expert @kobosm defends his PhD thesis on combining kernel density estimators. Good luck!

Yay! #CERMINE won the best performing approach award at the #ESWC2015 semantic publishing challenge! @d_tkaczyk @adalab_icm @ICM_UW

Semantic publishing challenge award goes to Poland. Gratulacje !! @bolikowski #eswc2015