Ontology-based knowledge integration framework: a systematic review

The key to the competitive edge of knowledge-based firms is Knowledge Integration (KI) for improved efficiency, effectiveness, and innovation. Despite its significance, KI faces substantial technical challenges, primarily due to the variety of knowledge representations that impede integration. To address these problems, Knowledge Integration Frameworks (KIFs) have been developed as structured systems or models that support integration, prevent duplication, and inform decision-making. This paper presents an overview of ontology-based KIFs for the last five years. According to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we identified 47 articles in Scopus and Web of Science (WoS). Consequently, we selected 18 high-quality studies based on strict inclusion/exclusion criteria. The survey discusses the construction and employment of ontologies within the knowledge inference frameworks. Although ontologies can provide precise semantics, existing platforms primarily rely on structured data within specific domains, which limits their applicability to unstructured or cross-domain contexts. Furthermore, a majority of them are stakeholder-excluded, unscalable, and lack uniform evaluation benchmarks. Most of the work is based on case studies with little consideration of quantitative benchmarks. Notably, 22% of the identified methods support partial automation in KI, 17% provide methods for stakeholder feedback, but none assess performance measures or comparisons. These findings suggest a disjunction between theoretical constructs and applied principles. This study contributes to the field by presenting limitations and future directions. There is a desire for designs applicable across multiple substitutable disciplines, benchmark datasets, and stakeholder-informed methods.

Text 124887.pdf - Published Version Available under License Creative Commons Attribution Share Alike. Download (834kB) Official URL or Download Paper: https://ijaseit.insightsociety.org/index.php/ijase...

Actions (login required)

View Item