A Governance Aware Ontology Based Analytics Framework for Scholarly Knowledge Graphs Using Budgeted Semantic Coverage Optimization

This study presents GAGER, a governance-aware ontology-based framework for scholarly knowledge graph analytics evaluated as a single-university case study. The framework integrates bibliometric normalization, semantic organization, thematic clustering, provenance control, governance validation, and budgeted semantic coverage optimization. It transforms Scopus-based institutional bibliographic records into an operational knowledge infrastructure in which documents, authors, institutions, countries, publication venues, document types, knowledge areas, thematic clusters, and Sustainable Development Goal tags are jointly represented through a traceable semantic graph. GAGER uses the ontology to define candidate semantic anchors and applies a maximum-coverage optimization model to select the most informative governance-valid anchors under explicit budget constraints. Experiments using a real single-university corpus show that the proposed governance-validated ontology-plus-optimization strategy achieves higher semantic coverage and aggregate decision quality than ontology-only selection, random selection, and frequency-based selection baselines evaluated under the same budget constraints. The governance layer also has a measurable operational effect, reducing the initial candidate-anchor space from 27,542 anchors to 8320 validated anchors before optimization. The contribution is therefore a methodologically reproducible single-institution analytical framework that connects semantic governance, provenance, graph-based bibliometric representation, and budgeted optimization to support institutional research analytics, thematic prioritization, and evidence-based decision-making within the analyzed university context.