Knowledge Tree
Relations (1)
related 2.81 — strongly supporting 6 facts
Quality assurance is a fundamental component of the knowledge graph construction lifecycle, as it is explicitly identified as a core requirement [1] and a cross-cutting concern that ensures ontological and data integrity [2]. Furthermore, it is integrated into every stage of the construction process [3] and is a critical feature in specific construction solutions like HKGB [4], NELL [5], and World KG [6].
Facts (6)
Sources
Construction of Knowledge Graphs: State and Challenges - arXiv arxiv.org 6 facts
claimThe NELL knowledge graph construction solution requires only final user approval of the correctness of extracted values or patterns for quality assurance.
claimThe World KG approach to knowledge graph construction manually verifies all matches to external ontologies for quality assurance.
claimThe HKGB knowledge graph construction solution relies heavily on user interaction for quality assurance.
claimRequirements for knowledge graph construction and maintenance are grouped into four aspects: input consumption, incremental data processing capabilities, tooling/pipelining, and quality assurance.
claimQuality assurance is necessary throughout the entire Knowledge Graph construction process, including source selection, data cleaning, knowledge extraction, ontology evolution, and entity fusion.
claimQuality assurance in knowledge graph construction is a cross-cutting topic that addresses ontological consistency, data quality of entities and relations (comprehensiveness), and domain coverage.