Article Preview
TopIntroduction
The World Wide Web is growing at a fast pace with lots of data available, but finding relevant information efficiently and timely is still a challenging job. Therefore, a layer of semantics is added to the current Web to make the machines understand the web documents (Berners-Lee, Hendler, & Lassila, 2001). The Semantic Web not only puts data on the Web but also establishes links among data so that humans and machines can explore the web of data rather than the web of documents and find other related data (Berners-Lee, 2006; Bizer, Heath, & Berners-Lee, 2009). This makes the Web more meaningful with the emergence of new paradigms such as Linked Open Data (LOD). Linked Open Data, an application of Semantic Web, can be defined as the Linked Data that is released under the open license and can be reused for free. Tim Berners-Lee presented a set of four basic rules for publishing, connecting and consuming the structured data on the Web (Berners-Lee, 2006). This Linked Data should have some general features including openness, modularity, scalability, and connectivity (Berners-Lee, 2009). These features make LOD as a source of knowledge discovery with several applications in multiple domains including e-government, health, education, and research.
Publishing data on LOD requires five basic steps including data preparation, dataset selection, plugging into the LOD cloud, its announcement, and identifying social contract (“Best Practices for Publishing Linked Data,” 2014; “Cookbook for Open Government Linked Data,” 2011; Hausenblas, 2009; Peinl, 2016). Figure 1 graphically illustrates these steps. First, the data is converted to triples. Second, the desired datasets (which could enrich the content) are selected (using the “follow your nose” principle). Third, the dataset is plugged into the LOD cloud for integrating data seamlessly to the chosen datasets to allow users and machines to consume it. Fourth, the data is published on a web server for its use and reuse. Finally, the implicit social contract with the consumers is recognized, which includes the responsibility of data availability, data freshness, accuracy, and maintenance (“Best Practices for Publishing Linked Data,” 2014; “Cookbook for Open Government Linked Data,” 2011; Hausenblas, 2009). The details of tools for the data extraction, transformation, storage, visualization and quality in publishing data on LOD can be found in the relevant literature (Bikakis, Tsinaraki, Gioldasis, Stavrakantonakis, & Christodoulakis, 2013; Butt, Haller, & Xie, 2015; Ferrara, Meo, Fiumara, & Baumgartner, 2014; Gangemi, 2013; Purohit et al., 2016).
Figure 1.
A typical LOD environment