Despite many e orts for making data about scholarly publications available on the Web of Data, information about academic conferences is still contained in (at best) free-text format. Availability of this data in a structured format would enable more e cient decision making for researchers, libraries, publishers, funding and evaluation bodies. This demo paper showcases the Springer Linked Open Data (LOD) conference portal (available at http://lod.springer.com). We cover the architecture, vocabularies and features of the portal and present usage scenarios.
Linked Open Data about conferences is often provided not directly by the
publishers, journals or conferences, but by third-parties, such as DBLP, COLINDA,
WikiCfP or EventSeer. A prominent exception is the Semantic Web Dog Food
repository1, now being relaunched as the Scholarly Data project [
This context led us to launching the Conference Linked Open Data portal,
available at http://lod.springer.com, on the 30th of March 2015 [
In the following, we list the main characteristics of the LOD portal. The data in the portal is freely available under CC0 1.0 Universal license.
Topical coverage and data sources. When launched in 2015 the dataset only contained conferences, conference series and proceedings in computer science. As of late 2016, the coverage was extended to include data about newly published conference proceedings in other disciplines, mainly engineering and mathematics. In early 2017 we have started adding recent books to the portal, so there are 57,043 books (10,093 of which are proceedings) and 1,203,912 book chapters and papers. In total, together with all the properties, the portal contains around 20 million triples. New information about the published conferences is loaded to the portal on a daily basis and monthly dumps are provided.
Interlinking and technologies. The conference series in the portal are
linked to the conference series in DBLP2. Thus, the portal respects the FAIR
principles3 and provides 5 star Linked Open Data [
Vocabularies. We use custom vocabularies as it makes mapping to our internal formats easier. However, in the future we would like to have mappings to the most common vocabularies, similarly to what Nature did in the past4.
The ontology les are included in the dumps at http://lod.springer. com/data/dumps. The complete description of the entities and their properties is available at http://lod.springer.com/wiki/bin/view/Linked+Open+ Data/Data+Description.
In the following, we list the main entities of the dataset: { conference series, which has a unique conference series ID, a conference series ID at DBLP, a conference series name, and links conferences from di erent years. For instance, International Semantic Web Conference, http: //lod.springer.com/data/html/confseries/semweb; { conference, which is characterized by a conference acronym, name, year, and number, the city and country, the start and end dates. The conference is also linked to the conference series and all books, which contain the proceedings of the conferences. For instance, ISWC 2016, http://lod.springer. com/data/html/conference/semweb2016; { book, which is characterized by usual bibliographic attributes, such as ISBN, DOI, editors; the acronym and name of the book series, the volume number, the title and subtitle. The book also contains information about whether it is indexed in Scopus and Ei Compendex and is linked to the conference and lists all the chapters (papers in the proceedings). See http://lod.springer. com/data/html/book/978-3-319-46523-4. 2 see, for instance, http://lod.springer.com/data/html/confseries/semweb 3 https://www.force11.org/group/fairgroup/fairprinciples 4 https://www.nature.com/ontologies/datasets/linksets/ { chapter, which has a DOI and a title, authors, the publication date and copyright year, page numbers. See http://lod.springer.com/data/html/ bookchapter/978-3-319-46523-4_1 for an example.
An important feature of the portal is that conference series together with conferences are treated as rst-class objects, similar to scienti c journals and issues. This addresses the fact that in computer science, conference series are comparable to journals regarding their impact and importance for publishing. 3
Conference names, dates, locations etc. have previously only been available as the unstructured subtitles of our proceedings volumes. During the development of the LOD conference portal this information has been transformed to a wellde ned and structured RDF format and integrated into the Springer internal work ow system. Figure 1 shows how the metadata about conferences ows through Springer systems: 1) conference-related information is entered during production; 2) it is stored in the product data; 3) it is included in the chapter and conference series pages on SpringerLink 5 as HTML-embedded microdata and 4) imported to the LOD portal as RDF.
Google Scholar uses the data about the conference where a speci c paper is published for improving Google Scholar Metrics6, a rating of publication venues in di erent disciplines. 5 see https://link.springer.com/conference/semweb, for an example of the ISWC conference series page 6 https://scholar.google.de/intl/en/scholar/metrics.html
Users can check if their conference papers published in proceedings are indexed in Scopus or Ei Compendex. Conference organizers or authors can search ISBN, DOI, conference acronym or the volume number of the book series and see if the proceedings are indexed or not (see Figure 2). Such services are enabled by the Scopus and Ei Compendex APIs.
Semantic web researchers can also access the portal via the SPARQL endpoint at http://lod.springer.com/sparql-form/index.html.