evidences of users' competence in the Linked Data Cloud
By now many datasets have been published in the Linking Open Data community project. Interlinked among each other, those datasets form an enormous and growing Web of Data that hosts much valuable information. Linked Data Cloud, maintained by Richard Cyganiak and Anja Jentzsch provides a visual overview of datasets in the cloud and their connections. The purpose of mapping the evidences of competence on the Linked Data Web is to extract a subset of Linked Data Sets that is relevant for determining a user's interests and competence in a particular domain; and eventually his/her capability to solve particular problems.
The Map of Evidence of Competence (MEC) provides a list of sources of evidence on user's (and organizational) competence. We don't limit the map only to direct evidences (like experience on a project, educational course, etc.), but we take into account the information that can be relevant to better assess the expertise of the user (e.g., social graph can be important for user ranking). The goal of creating MEC is to maintain a list of all relevant information sources that can be used to estimate competence on the Web, with a special focus on Social Web sources. We looked at Linked Data Sources in order to find which ones contain valuable information about evidence of competence, and as result we created MEC where relevant Linked Data Sources are grouped by the type of evidence they provide.
In order to provide better understanding of the map, we have spilled color on the Linked Data cloud to mark the type of evidence that is contained in each dataset (e.g., yellow color of (micro)blogs is spilled on SIOC Sites). As some datasets (e.g., FOAF Files) contain many types of evidence of competence, they reside on the crossroads of several colors. The subcloud on the right that is on green color represents datasets with general concepts that are not evidences of competence but contain useful information for identifying competence domains and similar meta-stuff.
Size of Linked Data Sets and their connections correspond to the original Linked Data Cloud, as the dataset visual representations are taken from the original work of Richard Cyganiak and Anja Jentzsch. As in the original LOD Cloud, the size of the circles corresponds to the number of triples in each dataset. The numbers are usually provided by the dataset publishers, and are sometimes rough estimates.
Circle size Triple count
Very large >1B
Large 1B-10M
Medium 10M-500k
Small 500k-10k
Very small <10k
The direction of the arrows indicate the dataset that contains the links, e.g., an arrow from A to B means that dataset A contains RDF triples that use identifiers from B. Bidirectional arrows usually indicate that the links are mirrored in both datasets. The thickness corresponds to the number of links.
Arrow thickness Triple count
Thick >100k
Medium 100k-1k
Thin <1k
You can help us improve the MEC We welcome your feedback as we would like to improve this Map and make it even more useful and complete. If you can think of any other source of evidence of competence, or any other type of evidence, do not hesitate to write to milan.stankovic@hypios.com .
Current version 0.1, created on 08/12/2009
This work by Milan Stankovic is licensed under Attribution-Share Alike 3.0 Unported Licence.
This research is a result of joint efforts of Hypios and LaLIC, Université Paris-Sorbonne Ackgnowledgement
The author whishes to thank Richard Cyganiak for verifying the correctness of the MEC with regard to the original LOD Cloud. Claudia Wagner also gave valuable comments and suggestions to improve the MEC.
