search

concepticon / concepticon-data

The curation repository for the data behind Concepticon.
https://concepticon.clld.org
32 stars 35 forks source link

GECO corpus on mono- and bilingual sentence reading #426

Closed LinguList closed 4 years ago

LinguList commented 6 years ago

If you look at this corpus, you'll see that it is an excellent example for metadata, as it provides a rather detailed annotation of a full novel, along with eyetracking data:

I imagine, one could automatically link the data as metadata, but it would require to transform the original dataset, as they have repeating words, so from these, we'll need to refererence back to lines in the original data.

There is a study of the team that suggests that AoA has an impact on bilingual sentence reading. This is extremely interesting, as it means: if you know a concept in your mother-tongue, you'll learn it faster in another language.

AnnikaTjuka commented 4 years ago

We would need to find a solution for multiple occurrences of a word. In addition, sentence reading and eye-tracking are not the focus in NoRaRe. Thus, I'll close this issue.