Semi-automatic staging area for high-quality structured data extraction from scientific literature
Résumé
In this study, we propose a staging-area for ingesting new superconductors experimental data in SuperCon that is machine-collected from scientific articles. Our objective is to enhance efficiency of updating SuperCon while maintaining or enhancing the data quality. We present a semi-automatic staging area driven by a workflow combining automatic and manual processes on the extracted database. An anomaly detection automatic process aims pre-screen the collected data. Users can then manually correct any errors through user interface tailored to simplify the data verification on the original PDF documents. Additionally, when a record is corrected, its raw data is collected and utilised to improve machine learning models as training data. Evaluation experiments demonstrates that our staging area significantly improves data quality with an increase of 40% in F1-score when comparing to the traditional manual approach of reading PDF documents and recording information in an Excel file. This improvement is primarily attributed to a reduction in missing or overlooked information, resulting in a 6% increase in precision and a 50% increase in recall.
Origine : Fichiers produits par l'(les) auteur(s)