Research corpus for case-based reasoning on regulatory documents, Version 1.0 alpha
Please be patient. The corpus is still in alpha condition. We are continously working on the corpus and files. The first stable release will be marked as 1.0 beta
Copyright
This is a RESEARCH BODY, NOT real guidance information from the Internatioal Atomic Energy Agency (IAEA). The corpus bases on publications from the IAEA. All rights for IAEA content belong to the IAEA. For direct information and guidance visit www.iaea.org
For questions on the corpus please write to a.korger[ät]angesagt-gmbh.de
Folders:
pyPIRI contains different pipeline components for regulatory knowledge management, complex semantic concept recognition, and relation extraction. All components use spaCy and webATHEN as fundamental technologies.
webATHEN contains type system descriptors to use the piri system with the webATHEN annotation environment.
Files:
PdfToText.java
Uses Apache PDFBox to convert all PDf-files in a folder to TXT-files.
CreateLuceneIndexFromFolder.java
Creates a Apache Lucene index from the input folder.
PIRIOntologyToCaseBase.java
Takes an ontology file as input and creates two CSV-files from it. One containing all information-unit-cases and one containing all PIRI-cases derivable from the input ontology.
PIRI-HighlightPDFToOntology-ontology.ttl
Ontology containing classes created from annotated PDF-files.
myCBR-model-PIRI-Fire-Version-1-0-beta.prj
Project-file for myCBR containing the case-based model.