공공데이터 재사용성 향상을 위한 9차원 품질 진단 프레임워크의 설계와 실증

초록

As the release of public open data expands, the need for systematic quality assessment to ensure reusability has grown. Beyond traditional indicators such as completeness and validity, the analysis, linkage, and repurposing of such data require multi-dimensional evaluation encompassing machine-readability and reference-code interoperability. This study proposes a nine-dimensional framework for the automated quality diagnosis of public open data CSV files. Extending four quality indicators of the Ministry of the Interior and Safety and the ISO/IEC 25024 metrics, it operationalizes reusability and interoperability into measurable form and adds five dimensions: structural integrity, schema stability, semantic redundancy, reusability, and standard compliance. A fully automated diagnostic pipeline is implemented through a gate mechanism and a reference data-based rule engine. Analysis of 1,258 CSV datasets from 19 institutions under the Ministry of Trade, Industry and Energy yielded an average score of 77.7 (Grade B). Timeliness (D4) emerged as the most critical issue with an achievement rate of 16.2%, while structural integrity (D5, 96.7%) and semantic redundancy (D7, 98.4%) were satisfactory. Institutional averages ranged from 63.0 to 86.3, confirming substantial variance and the need for institution-tailored curation beyond ministry-level guidelines.

키워드