Akademik Veri Yönetim Sistemi
IEEE Access, cilt.11, ss.59226-59237, 2023 (SCI-Expanded)
Although Knowledge Graphs (KGs) are widely used, they suffer from hosting false information. To eliminate this deficiency, different studies have been conducted to correct and complete them. These studies correct triples, relations, relation types, and literal values or enrich the KG by generating new triples and relations. The proposed methods can be grouped as closed-world approaches that take into account the KG itself or open-world approaches using external resources. Recent studies also take into account the confidence of triples for the refinement process. The confidence values calculated in these studies affect either the triple itself or the ground rule for rule-based models. In this study, a propagation approach based on the confidence of triples is proposed for the refinement process. This method ensures that the effect of confidence spreads over the KG without being limited to a single triple. This makes the KG continuously more stable by further strengthening strong relations and eliminating weak ones. Another limitation of existing studies is that they treat refinement as a one-time process and do not provide due importance to process performance. However, real-world KGs are live, dynamic, and constantly evolving systems. Therefore, the proposed approach should support continuous refinement. To measure this, experiments were conducted with varying data sizes and false triple rates. The experiments were performed using the FB15K, NELL, WN18, and YAGO3-10 datasets, which are commonly used in refinement studies. Despite the increase in data size and false information rate, an average accuracy of 90% and an average precision of 98% was achieved across all datasets.