The Web Conference 2018 took place from 23 to 27 April 2018. This and the next post are an account of the contributions made by WDAqua.
The Web Conference took place from the 23rd to the 27th of April in Lyon. It was a massive event, with more than 2000 attendees, 1172 paper submissions to the research track, 27 workshops, 30 demos, and more.
The conference was very competitive: the acceptance rate in the research track was 15% (A+ rated). WDAqua had a strong presence, with 1 full paper presentation, 2 workshop papers, and 1 workshop jointly organised by some of the members of our consortium.
Data search was one of the main topics at the conference. Natasha Noy from Google pointed out the importance of understanding how users search for data and how to detect data-related queries, during her keynote at the SAVE-SD workshop. This was also one of the topics of the PROFILE & DATA:SEARCH workshop, organised by our Laura Koesten, Elena Demidova, Stefan Dietze, and Elena Simperl, plus Vadim Savenkov, John Breslin, and Oscar Corcho. The workshop aimed to look at different aspects of data search and retrieval on the Web, bringing together researchers and practitioners from different disciplines, such as information retrieval, the semantic web, and human-computer interaction. The two keynotes hosted by the workshop reflected this diversity of perspectives. Maarten de Rijke, from the University of Amsterdam, presented the work done by his team to develop optimal algorithms for data search and retrieval. On the other hand, Aidan Hogan talked about the use of graphs in data search and the importance of profiling to make these usable.
The workshop included a paper presentation from one WDAqua student, Emilia Kacprzak. Her work focused on understanding how people write queries to look for data and what is important in data requests. The paper can be found here.
Listening to the best panel with @JeniT @stefandietze @pgroth , Natasha Noy and Aidan Hogan answering the question “Do we need a Google for data?” (and how could it look like?) at our Profiling & Data Search workshop @TheWebConf #datasearch pic.twitter.com/eFxtfcV9XL— laura (@laurakoesten) April 24, 2018
The level of the panel was exceptionally good. Its members – Paul Groth, Aidan Hogan, Jeni Tennison, Stefan Dietze and Natasha Noy) – debated the question “do we need a Google for data?”. The answers were quite nuanced: although the participants agreed with the need to have powerful search instruments for data, their perspectives differed in terms of which points they stressed the most, e.g. valuing the importance of using currently existing tools or having domain-specific search portals. For those who missed it at The Web Conference, a new edition of the DATA:SEARCH workshop has already been already planned for SIGIR’18.
Members of WDAqua took part also in other workshops. Elena Simperl was one of the keynote speakers at the HUML workshop. She presented the work done with her supervisees Alessandro Piscopo and Lucie Kaffee about Wikidata’s social fabric and the effect of human and automated contributions on its data quality.
Keynote #2: @esimperl @unisouthampton on "Loops of humans and bots in Wikidata" at #HumL #workshop "Augmenting Intelligence with Humans-in-the-Loop" @HumLworkshop at @www2018 #TheWebConf pic.twitter.com/cPdusIhUVa— Lora Aroyo (@laroyo) April 24, 2018
Last but not least, Lucie presented her poster at the Wiki Workshop. In her study, she shed light about the stability of human-readable labels of Wikidata over time.