Semantic tier Sample Clauses

Semantic tier. The semantic tier of BOUNCE will provide a unified, homogenised view over the underlying data sources. A detailed view of the Semantic tier is shown in Figure 5 and consists of the following components: the data lake, the data cleaning tools, the cleaned databases, the semantic integration tool and the APIs. Below we will describe each one of those components in detail.  Data Lake: At the bottom layer, all inserted data will be staged in a repository with multiple databases forming a data lake. Data will include the anonymized retrospective and prospective data provided by HUS, IEO, HUJ and CHAMP and imported data from external registries. The data will reside in the data lake in their original formatting to be further cleaned and processed. The data residing in the data lake will then be cleaned using appropriate data cleaning tools. Cleaning will entail missing value handling, identification of erroneous records etc. The cleaned data, will again be staged in the data lake.  Semantic Integration & Mapping: Using the semantic model, i.e. the BOUNCE ontology that will be constructed within the BOUNCE project (WP3), all available data will be modelled, mapped and semantically uplifted to triples. There will be available the possibility to perform ETL, exporting selected data to an RDF triple store, or of the real time integration through a virtual data source.  Data Access APIs: All data, both integrated and those residing at the data lake will be accessible through various APIs that will respect all necessary security and confidentiality requirements established by the security layer of the BOUNCE project.