cLODg implements a methodology to produce Linked Data to describe a scientific conference and its publications, participants and events. To generate conference metadata we assume that you have initial data available (with some specific format). We then use use D2R conversion rules to produce metadata described with Conference Ontology. While we are in the process of creating a single-click process, we describe here the current workflow to follow.
The workflow consists of:
We currently accept data in the following formats:
- rdf data: any conference metadata represented with the Semantic Web Conference Ontology
- example of ESWC2015 data from the SWDF dataset
- csv files exported from EasyChair
- samples of ESWC2016 data exported from easychair about paper authors, program committee, submitted papers and conference tracks (Provided data is simplified for privacy reasons)
- csv describing keynote, organizing committee, sessions, talks and special events
- samples of data from ESWC2016 about organizing committee, program committee, keynote talks, conference sessions and and conference talks
- we still support conversion from XML EasyChair output, although this is currently not provided anymore by easychair (XML schema)
We generate data according to the Conference Ontology. The Conference Ontology is aligned to a set of other ontologies/vocabularies that allows interoperability with other linked datasets. Additionally, the Conference Ontology is aligned to the Semantic Web Conference Ontology for backward compatibility with the Semantic Web Dog Food dataset.
Nevertheless, we are in the process of replacing MySQL with HSQLDB to support a fully integrated Java process (to facilitate data creation for the end user as a single process with cLODg). Code for this will be published ASAP.
We implemented a Web based application to collect user feedback. This is an optional step that a conference can implement to send generated metadata (in a user-friendly format) to people involved and ask them to verify them.