I've noticed lately that the Merge command regularly returns invalid hits, particularly so in morphological searches. This is the case with the search set to either article or paragraph. Rather than this older Merge approach to utilizing the CATSS database, please consider the added benefit that morphological tagging would bring to tools. Enabling tools for a searchable field, of which one could be morphological tagging, would permit comparative searches on article which would be far more accurate than the current Merge work around. Regardless, please fix the Merge command. Thanks.
Posted 24 July 2014 - 03:07 PM
Can you give some examples of incorrect results?
By day: Consultant for Oaktree
By night: Freelance Trombonist and Private Instructor
Posted 24 July 2014 - 03:16 PM
Nearly any morphological search will contain errors. E.g., run a search of verb nifal in HMT with the range command argument of Jeremiah. The first occurrence is wrong, it's not a nifal (and it is correctly tagged as a qal in HMT/BHS-T). Yet, MT - LXX considers it a match, and I would bet this example digs down to the core problem with the Merge Command.
Posted 25 July 2014 - 02:55 PM
There is no way that we can invest in tagging this tool.
The MERGE search is not accurate especially for morphological searches, since it looks in the hit set of verses for all the inflected form hits. It's even worse for phrases or constructs as if there is a common word in the phrase it will show up as a hit in many verses where it is not part of the phrase at all. This is simply a limitation of the tool and the feature, and there is no easy workaround. You must filter the results yourself.
Posted 26 July 2014 - 09:46 AM
Thanks for the answer. It's unfortunate that an edited Merge command or some other advanced feature will not be considered.
Edited by James Tucker, 26 July 2014 - 09:47 AM.
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users