Home Forums Software ELAN Interlinearization mode

Interlinearization mode

This topic contains 3 replies, has 2 voices, and was last updated by  Han 3 months, 1 week ago.

Viewing 4 posts - 1 through 4 (of 4 total)
Author Posts
Author Posts
March 8, 2019 at 11:18 #12797

Jo

My colleague and I are having some issues when using Interlinearization mode. One issue is that the ‘include variants in the parsing process’ setting is not working – the analyzers don’t gloss for the variants in the lexicon despite this setting being activated. Are there known issues with this function?

March 11, 2019 at 11:52 #12798

Han

There are still many things to improve in the interlinearization mode. I’m not sure if this was already a known issue, but I can reproduce the problem, at least some of the times.
I haven’t found out yet exactly when it seems to work and when not. One pitfall is that it is easy to forget to click Apply Settings after (un)ticking the checkbox.
Then, when the setting does have an effect, it seems the variant is used in the matching process, but when a match is found, the main field (lexical-unit) is returned in the output, in the suggestions. I don’t know if this is the intended behaviour.
In your case, is the variant not used at all, not even after changing the setting and maybe closing and reopening the file? Or is the problem that the variant doesn’t show up in the suggestions?

-Han

March 12, 2019 at 15:33 #12802

Jo

Hi Han,

Thanks for your response. The problem I’m having is that the variant does not appear to be being used by the analyzer at all. I’m using a Gloss analyzer to check (in the lexicon I’ve imported into Elan) for the lexical units on my ‘Words’ tier, and input the glosses from the lexicon on my ‘Gloss’ tier. I’m using the ‘Include variants…’ setting because the transcriptions contain alternative forms of some lexical units and I would like the analyzer to also check for these in the lexicon (where they are included as Variant) and input the appropriate gloss. Is the setting perhaps not designed to be used by a Gloss analyzer in this way?

March 13, 2019 at 10:24 #12804

Han

Ah, good that you mention that. The Gloss analyzer currently indeed doesn’t support look-up based on any other field than lexical-unit. The configuration panel for the Gloss analyzer doesn’t have a checkbox for that either.
I looked into the code and there is a “to do” comment there that this should be made more flexible.

I guess you can’t use the Lexicon analyzer instead because its configuration still assumes a tier dependency between the first target (e.g. morph) and the second (e.g. gloss).

Viewing 4 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic.