Appendix D: System generated clusters
Name | Description |
|---|---|
"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 |