process for manual classification of committee protocol parts using Catma

[OriHoch]

catma allows to manually clasify parts of committee protocols

Bar Ilan University (BIU) uses it to classify parts of the protocol according to the following tags:

• judicial decision
• constitutional turns (turns == parts)
• anticipating judicial review
• doubt (with optional comment)

In Catma, documents are original committee protocol files (uploaded directly by BIU staff), which Catma parses to pure text. Each document can contain multiple annotations by different users (currently, in BIU, a single user annotates each document)

To parse results of the classification and integrate back to knesset data:

• export the annotations from Catma (manually)
• parse the results (using some Python code which BIU has)
• relate the results to the existing protocol parts - should be done by text matching
• integrate with #178 - to include this data for each part's classifications

[OriHoch]

code for parsing of the results: show notebook