Linked Data: Structured data on the Web
David Wood, Marsha Zaidman, Luke Ruth, Michael Hausenblas
Format: PDF / Kindle (mobi) / ePub
The current Web is mostly a collection of linked documents useful for human consumption. The evolving Web includes data collections that may be identified and linked so that they can be consumed by automated processes. The W3C approach to this is Linked Data and it is already used by Google, Facebook, IBM, Oracle, and government agencies worldwide.
Written to be immediately useful to Web developers, this book requires no previous exposure to Linked Data or Semantic Web technologies.
calling client. See also request. REST See Representational State Transfer. REPRESENTATIONAL STATE TRANSFER An architectural style for information systems used on the Web. It explains some of the Web’s key features, such as extreme scalability and robustness to change. REST API An API implemented using HTTP and the principles of REST to allow actions on web resources. The most common actions are to create, retrieve, update, and delete resources. SEMANTIC WEB An evolution or part of the
domain you do own. Making URIs (also known colloqui- ally as “minting” URIs) in someone else’s DNS domain is considered extremely rude. Although it’s true that RDF allows anyone to say anything about anything (just as the Web does), it’s also true that you need to be careful about sources of data (the same as for sources of documents on the Web). Minting URIs that use your own domain and publishing them on your own domain help people to trust your data. This is part of the social contract
Data 71 Figure 3.14 Illustration of linked relationships of common products Jay Myers of Best Buy reports that as many as 100 different criteria could affect the purchase of one product.2 He expects that the use of semantic data can enhance the Best Buy site, improve the visibility of more than 85% of the products, and help con- sumers identify more appropriate products. Figure 3.14 demonstrates the semantic relationships between products. Myers expects that “RDFa can ultimately create
24649 24651 24656 24657 24658 We’ll store our transformed and aggregated data using a Fuseki RDF database from the Apache Jena project. Fuseki includes a lightweight, in-memory database perfect for working with small amounts of RDF data. In order to execute our application, you’ll need to install Fuseki and set permissions by following the instructions at http:// jena.apache.org/documentation/serving_data/index.html. You should already have Python installed from chapter 4. Our
objects may be a full URI, a blank node, or a literal. This format will be adopted as a standard in the latter part of 2013. 5 To modify an existing dataset, follow the steps outlined in step 4, but modify an existing .nt file or create a separate one. 6 Finalize your edits by sending a pull request via GitHub; see figure 8.13. This operation requests that DBpedia add your triples to their repository. Figure 8.13 GitHub screenshot of Pull Request option NOTE All information uploaded to