Abstract

Document management systems (DMS) are widely used for the management of business documents because they use metadata to organise and categorise digital documents. However, they are often based on unstructured and monolithic files and this structure raises questions about the quality and completeness of the information. To overcome this problem, this paper proposes a semantic rulebased approach for an RDF-based DMS, which uses a combination of ontology and Shapes Constraint Language (SHACL) rules to integrate legal aspects, validate data (expressed in an RDF triple store), reason and infer new information. The process is dynamic because the proposed DMS can automatically reason and create new inferences based on the information and data extracted from documents. The process also reasons on user profiles and underlying rules, capturing specific legal regulations, enabling further accurate and automated document management. The ontology used in the process captures specific concepts of Swiss tax returns, while the SHACL rules serve to reason about actual RDF triples relating to different tax households. The proposed DMS is innovative for its ability to reason on a specific domain, improve the accuracy and completeness of information managed. This work is relevant for any domain involving administrative documents and regulations (e.g. fiduciary or insurance sector).

Authors

Cappelli, Maria Assunta; Caselli, Ashley; Di Marzo Serugendo, Giovanna.

URL

https://ksiresearch.org/jvlc/journal/JVLC2023N2/paper034.pdf