Acquisition

From ONTORULE Show Case

Within the acquisition phase the business expert Gary and the business analyst Marc develop a formal, validated and complete business model.

For extraction relevant information the business analyst takes all documented input into account. This includes textual documents in natural language, such as business policies, laws, regulations, treaties, protocols, contracts, but also spread sheets and presentation slides as well as all kind of diagrams. Additionally the experience and knowledge of the business expert Gary is taken into account.

Within ONTORULE we distinguish between two types of knowledge. On the one hand the business defines it's vocabulary to communicate in a consistent way. The knowledge can be extracted from various documents and from the business experts. On the other hand the business defines business policies from which business rules can be derived. The knowledge can be found in policy documents or gathered from business experts.

In order to extract the knowledge from various sources ONTORULE provides tools and methodologies based on natural language extraction as well as on established acquisition and modelling paradigms.

Associated Abstract Components

	Component Description
Candidate rule extraction from NL	This component aids to extract candidate rules from pieces of text documents. (X)HTML; OpenDocument Format and Office Open XML are preferred as the input format for text documents.
Modeling	This component assists business analysts during the modeling process to produce valid and verified BPMN+SBVR specifications of business models.
Ontology extraction from NL	This component aids to extract OWL2 ontologies and SKOS linguistic knowledge from pieces of text documents. (X)HTML; OpenDocument Format and Office Open XML are preferred as the input format for text documents.
Projection	This component supports the partial transformation of BPMN+SBVR business models into either OWL2; RIF or a combination of both.
Text annotation	This component permits to associate linguistic knowledge; artifacts of a (possibly partial) business model or elements of an OWL2 and RIF combination to fragments of text documents producing TELIX-annotated documents.
Text querying	This component retrieves text fragments; ontological elements or candidate rules in response business queries. The queries are expressed in SPARQL. In practice the goal is to retrieve the textual; ontological or rule elements that are linked to a combination of semantics elements (ontological elements or candidate rules).