Leading S in text

The front end.
Post Reply
userFFF
Posts: 2
Joined: Wed Oct 17, 2018 10:26 am

Leading S in text

Post by userFFF » Thu Jan 10, 2019 8:01 am

Hi. I am testing the tool with some samples. I got this strange behaviour: in the tornado report I see (some) text strings reported with a leading "S", for example (see attached report) "nature of injury" = "SLaceration" instead of "Laceration". "Burn" in the second item of the report is reported correctly instead.
Data is read from a CSV file and looks fine to me. I tried to find suggestions in the documentation by so far I did't find anything. I am confused: is this due to global setting? Or is it a problem in data reading?
Thank you very much for any suggestion.
Attachments
report.png
report.png (118.41 KiB) Viewed 500 times

shooltz[BayesFusion]
Site Admin
Posts: 1202
Joined: Mon Nov 26, 2007 5:51 pm

Re: Leading S in text

Post by shooltz[BayesFusion] » Thu Jan 10, 2019 11:52 am

The leading S is added by GeNIe when you learn network from a data file, and the value in the data column contains non-alphanumeric characters and underscores, or starts with non-letter. I believe your data file had a NATURE column with "aceration open wound" value. The spaces between the words caused GeNIe to convert it (including the "S" prefix). The same applies to "Foreign body", which was converted to SForeign_body

You can modify the loaded data (before learning) replace the values with spaces in GeNIe - use Edit|Replace option when data file spreadsheet is active.

Post Reply