Authors: Karol Rástočný
Tags: 2012, conceptual modeling, Mária Bieliková
A model of web information system content is crucial for its effective manipulation. We employ knowledge tags – metadata that describe an aspect of an information artifact for purpose of the modeling. Knowledge tags provide a lightweight semantics over the content, which serves web information systems also for sharing knowledge about the content and interconnections between information artifacts. Knowledge tags represent not only content based metadata, but also a collaboration metadata, e.g. aggregations of an implicit user feedback including interactions with the content. To allow this type of metadata we should provide means for knowledge tags repository providing flexible and fast access for effective reasoning. In this paper we address issues related to knowledge tags repository and its automatic maintenance. Main design issues are based on considering dynamic character of the web of information artifacts, which results in content changes in time that can invalidate knowledge tags. We realized the web-scale repository via the MongoDB database. Proposed repository stores knowledge tags in Open Annotation model and supports inference via distributed SPARQL query processing algorithm for MapReduce.Read the full paper here: https://link.springer.com/chapter/10.1007/978-3-642-33999-8_39
