Skip to main content

Reveal Review Publication

Appendix D: System generated clusters



"Not clusterable"

Documents which do not have text vectors

"Incoherent cluster"

Documents were detected with different topics in nature

"Delivery reports"

Documents which were detected as Email Delivery Report

"Short docs cluster"

Documents with short content and difficult to summarize

"Assumed summaries and reports"

Documents that are (or are similar to) tables, spreadsheets or reports

“Processing timeout”

Documents with processing timeout exception and also no topic found from their text vectors