The LiveData Data Model

LiveData is a catalog of catalogs. Each catalog, accessible through the LiveData web portal, collects and distributes data conforming to the DataScientia data model. Such a model consisits in the creation, management and distribution of the so-called "Diversity Aware Data". The key idea behind diversity aware data is to enhance the power of representation of the data, by emphasizing and making concrete the diversity of the data, at different levels. In fact, the information carried by data is not only represented by a single dataset. Rather, it is implicitly hidden in such a dataset by the data source. In the diversity aware data model the information carried by a dataset is represented as a composition of real world entities, described by using a specific language, having a data schema following a specific ontology, as well as using high quality standard data values. More concretely, a diversity aware data can be seen as the composition of 3 main elements; (i) a dataset provided by a data source, (ii) a language datasets describing the terminology adopted by the dataset, (iii) and an ontology representing the data scheme through which the information entities are linked each other. Moreover, the diversity aware data can be represented as a single object, shaped as a Knowledge Graph (KG) where the above three components are composed together. The diversity aware data produced and distributed by DataScientia, through the LiveData network, defines a new generation of data, where the interoperability, quality and data reuse, are the pillars for a more efficient, and less expensive, digital innovation.

---

       

Source Data: datasets, provided by data sources (external respect to the LiveData network) that have been cleaned and formatted by adopting well-known standards. Such datasets carry the data values to be associated with the entity according to the diversity aware data model.

---

       

Language Data: datasets, containing the definition and representation of the concepts used to express the information carried by the other datasets (like source data, ontologies and graph data).

---

       

Schema Data: ontologies which make explicit, and machine-readable, the ontological model of the information as it is implicitly represented in the standardized datasets. Such ontology data are used to model the structure of the entity, in the diversity aware data model, as well as the interaction among them.

---

       

Graph Data: Knowledge Graphs (KGs) designed to contain all the types of datasets described above, thus making the different levels of diversity explicit in a single object.

More detailed information about the LiveData data model can be found in the dedicated paper:
LiveData - A Worldwide Data Mesh for Stratified Data