ico

BootOX Help Page

You are here: Home » Training programme » Module: Bootstrapper » BootOX Help Page

This help page explains the purpose of each input parameter of BootOX.

Available schemata

The drop down shows the added data sources. Select the one to be bootstrapped.

Select Bootstrapping level

Select what the bootstrapper is given access to. The schema alone is enough to give a reasonable result. Adding the data will improve the result, but it makes the bootstrapping process more time consuming.

  • Schema driven (default): The bootstrapper only use the database schema.
  • Schema and data driven: The bootstrapper use the data in addition to the database schema. The selection of this option will enhance the bootstrapping process and a richer class hierarchy may be bootstrapped. For example, the table ‘Well’ may include a column storing the type of a concrete ‘Well’ (e.g. Dry, Gas, etc.), and by exploiting the content of the database one could bootstrap subclasses of ‘Well’ (e.g. ‘GasWell’ , ‘DryWell’).

Select Expressiveness for Bootstrapped Ontology

Select the expressiveness of the bootstrapped ontology. More expressive ontologies are able to express more complex statements, but they make reasoning harder.
A full description of the OWL 2 and its profiles can be found here.

  • OWL 2 QL (default):
    A fragment/sublanguage of OWL 2 which is suitable for applications where query answering is the most important reasoning task and the data is stored in a standard relational database system can be queried through an ontology via a simple rewriting mechanism, i.e., by rewriting the query into an SQL query that is then answered by the RDBMS system.
    Optique is such an application, hence this is the default option.
  • OWL 2 EL: This fragment/sublanguage of OWL 2 is particularly useful in applications employing ontologies that contain very large numbers of properties and/or classes (e.g. in biomedicine).
  • OWL 2 RL: This fragment/sublanguage is aimed at applications that require scalable reasoning without sacrificing too much expressive power. It is also suitable for querying large volumes of data as RDF triples.
  • OWL 2: The full OWL 2 profile. Using the whole OWL 2 expressiveness may not be suitable depending on the practical applications, but there are many off-the-shelf systems that support different OWL 2 reasoning tasks.

Select Constraint Schema for Bootstrapped Ontology

Select which kind of constraints to allow.

  • Global constraint schema: Select this to add global constraints, like functionality of roles and domain/range axioms, if they are allowed in the selected ontology expressiveness.
  • Local and global constraint schema: Select this to add both local and global constraints that fall into the selected ontology expressiveness.
  • Local constraint schema (default): Select this to add the most intuitive axioms and local constraints that fall into the selected ontology expressiveness.
    (*) If the bootstrapped ontology is to be integrated with domain ontologies, a local constraint schema is preferable to avoid unintended logical consequences. For example, in a concrete database/bootstrapped_ontology may be valid to say that the domain of the attribute “name” can only be ‘Company’, but other ontologies this attribute can also be used in other scopes (e.g., ‘Person’).

Select Attribute Naming Schema for Bootstrapped Ontology

Select how URIs for similar (same name and range) attributes are generated.

  • Creation of reusable mapping names: Generate the same URI for attributes with the same name and range (e.g. a case for attribute “name” which typically appears is more than one table).
  • Creation of unique mapping names: Generate unique URIs for each attribute.

Provenance mappings

Select how to create provenance mappings, which are additional (virtual) triples which explicitly describe the provenance of the data (e.g. database, schema, table, column)

  • At URI level: the provenance is added for each generated URI
  • At triple level: the provenance is added for each (virtual) triple via reification. The number of additional mappings may be substantial.

Additional bootstrapping options

  • Extract annotations for query formulation interface: Select this if you intend to use the QFI to construct queries over the bootstrapped ontology.
  • Extract schema-driven subclass relationships: It heuristically checks that if the primary key of a table A is also a foreign key to table B then A subclassof B.
  • Bootstrap disjointness axioms: Defines A and B to be disjoint if there is no subclass relationship (incl. transitive closure?) between them.
  • Identify empty tables and annotate bootstrapped classes accordingly: It builds a cleaner ontology and avoid showing in the VQS classes that does not contain any data.

Import domain ontology

BootOX is able to import existing domain ontologies.

  • Via ontology alignment: it aligns the bootstrapped ontology with the domain ontology using the system LogMap.
  • Via direct connection to the database: it bootstraps (R2RML) mappings referencing the domain ontology vocabulary. In this case it does not bootstrap an ontology.

OWL 2 QL approximation

Select this option to let BootOX approximate the ontology to fit into the OWL 2 QL profile. This is useful in the case where the aligned ontology is more expressive than OWL 2 QL.

Ontology and Mapping Storage

Select what to bootstrap, and what to name the assets.

  • Store Bootstrapped Ontology: Select to save the bootstrapped ontology. Give it a suitable name in the given input field.
  • Store Bootstrapped Mappings: Select to save the bootstrapped mappings. Give it a suitable name in the given input field. For large number of mappings, storage may be time consuming.
Show Comments