Towards observing, understanding and shaping the Web of Data
Linked Data Topic Profiles: provide structured information about the topics covered by Linked Datasets in the Linked Open Data Cloud. The topics represent DBpedia categories. For each dataset the topics are weighed to reflect their importance for a particular dataset. Additional information such as the DBpedia entities that support such a topic and their corresponding resource instances from the dataset are captured and interlinked through the Vocabulary of Links Schema. The data is accessible under the following SPARQL endpoint, or through the explorative search interface.
Billion Triples Challenge Dataset: the BTC Datasets provides crawls of a number of structured data sources over an extended period of time and opens the opportunity to analyse the temporal evolution of Web datasets over time.
LinkedUp Data Catalog: the LinkedUp data catalog is a repository/catalogue of Web datasets relevant to educational applications. It is provided according to the standard of the Web of Data, and is constructed based on input from the LinkedUp Community. It is a general-purpose resource for anyone interested in building applications for learning and education but especially meant to support participants to the LinkedUp Challenge. Mappings between resource types facilitate cross-dataset queries via joint SPARQL endpoint.
DyLDO: is an approach which crawls and analyses the LOD cloud focusing mostly on the dynamics of changes in datasets (predicates, number of instances, etc). The selection of documents or resources for analysis from the LOD cloud is determined by computing the PageRank score for resources part of the Billion Triples Challenge Dataset (BTC).