KP-LAB Knowledge Practices Laboratory -- Specification of the SWKM Knowledge Evolution, Recommendation, and Mining services

deliverablesThis deliverable presents the deep-level specification for the second release (M24) of the components responsible for advanced manipulation with the knowledge stored in the KP-Lab Semantic Web Knowledge Middleware (SWKM). The two components were defined in [D5.1] as Knowledge Mediator and Knowledge Matchmaker. The Knowledge Mediator services (change, comparison, versioning and registry) aim at providing functionalities to support evolving ontologies and RDF Knowledge Bases (KBs). Upon a change request, the change service will automatically determine the effects and side-effects of the request and present it to the caller for validation. A comparison service is necessary to allow one to compare two versions of an ontology or RDF KB and identify their differences. The above functionalities are coupled with a versioning system, which is used to make different versions of the same ontology (or RDF KB) persistent, and with the registry service, which allows the user to classify the stored ontologies, using some related metadata for easy access and manipulation. The Knowledge Matchmaker supports advanced mining and notification services for knowledge artefacts. It essentially enables to cluster/classify available information resources with respect to the employed ontologies, as well as, to notify about changes to content items produced/consumed within a group of learners according to explicitly subscribed preferences [DoWB]. The Notification service supports access to the knowledge repository for KP-Lab users (i.e. individual human users as well as various tools or software components) by keeping them aware of changes. Users will be able to subscribe their preferences to the KP-Lab system in order to be notified about the changes in the knowledge repository. Events (modifications) in the repository are matched with the subscriptions and notifications are propagated automatically to the users. Text Mining services are used to assist users when creating or updating the semantic descriptions of KP-Lab knowledge artefacts. The Classification Service, after a software-training period, will classify the artefacts under some pre-defined set of categories (e.g., ontology concepts), resulting in a semi-automatic generation of semantic descriptions. The Clustering Service will look for clusters of similar artefacts and automatically acquire conceptual maps from knowledge artefacts. This can lead to the update or even the creation of (new) KP-Lab ontologies managed in the sequel by the Knowledge Mediator. The services are described along with the proposed functionality for each one, based upon the motivating scenarios and the subsequent functional requirements. The functionality of the services is presented from the end-user perspective and divided into parts that form the major components of the SWKM knowledge evolution, recommendation and mining services architecture