A while back I looked at the submissions for DHBenelux 2016 in a blogpost, and I promised to compare to the two earlier conferences. Today is finally the day I publish these comparisons, looking at DHBenelux from 2014 to 2016.

Numbers and types of submissions

First, let’s look at the number of submissions for each conference. While 2014 received 61 submissions, this was 80 for 2015 and 116 for 2016^[1]Thanks to Marijn Koolen for the data for 2014 & 2015. As such, the third instalment is almost a doubling of the first. While the first two conferences separated posters and demos, and gave every talk the same status of ‘oral presentation’, this year’s conference has a single poster/demo category, and separates long and short papers. Still, for insight’s sake, here are the numbers:

• 2014:
  • 11 demo
  • 27 oral
  • 18 poster
  • 21 undefined^[2]This is a reflection of the two year old EasyChair data I received, rather than a reflection of the actual conference
• 2015:
  • 31 demo
  • 70 oral
  • 3 poster
• 2016:
  • 34 long paper
  • 55 short paper
  • 36 poster or demo

What’s more, in 2014 a total of two papers were rejected (97% acceptance), in 2015 four papers were rejected (95% acceptance), and in 2016 19 papers were rejected (84% acceptance), and another 17 submissions have been asked to change a paper to a poster.

Authors

As with the submissions, the number of authors submitting papers has grown over the years. In 2014 a total number of 113 authors wrote papers, which grew to 158 authors in 2015, and 204 in 2016. Since the Benelux isn’t that big, I expected that there would be a large overlap between the conferences, with a vast group of authors submitting every year. This is however not the case: a total of 406 unique authors submitted to the three conferences. The overlap is as follows:

• From 2014 to 2015,  28 authors returned (25% of 2014)
• From 2015 to 2016, 29 authors returned (18% of 2015)
• Another 12 authors submitted in 2014 and 2016, but not 2015 (11% of 2014)
• Finally, only 9 authors submitted papers in all three conferences (2% of total)^[3]Guess who is one of the nine? :)

Keywords

Finally, do we see evolution of the topics that are discussed at the conferences? I processed all the abstracts I could find through Voyant to get a sense of which words are used.

For 2014, I could do this for 57 abstracts (missing four), which contained 19k words and 4k unique words^[4]These are not exact figures, since many abstracts contain the paper title, author names, and references, but some do not. Moreover, I did not combine similar words in this analysis. This review is just to get a sense of the submissions. The most used words with 40 or more occurrences are in the below chart:

For 2015, I could do this for 75 abstracts (missing five), which contained 28k words and 6k unique words. See below chart for the most used words with 40 or more occurrences:

For 2016, I’ll repeat the results from the previous blogpost here for comparison: the 116 abstracts together contain 105k words, and 15.8k unique words. See below the chart with all words from the submission abstracts with over 100 occurrences.

For all three conferences, digital remains the keyword of choice. In 2014 and 2016, Dutch is fairly common, but not so much in 2015. An interesting word that comes up with all three conferences is new.

Wrap-up

That’s it for now. Overall we can see that DHBenelux is growing steadily, as does the global DH conference. With this the number of rejections has also increased this year, which could change the nature of the conference somewhat from a very inclusive ‘anyone is welcome to present’ to a more selective conference. Finally, despite being called “DHBenelux”, the pool of authors is much larger than I expected, and only a small portion of each conference is from authors who also submitted to the other conferences. All this is good news, as it shows the conference is definitely not a clique of people who know one another, but actually a varied and growing community. 

If you want more information on the submissions, ask away in the comments.