Performance (F1 score (%)) of the fine-tuned PLMs and ensemble models on closely related language combinations. The average weighted F1 with 95% confidence intervals (CIs).

Performance (F1 score (%)) of the fine-tuned PLMs and ensemble models on closely related language combinations. The average weighted F1 with 95% confidence intervals (CIs).

<p>Performance (F1 score (%)) of the fine-tuned PLMs and ensemble models on closely related language combinations. The average weighted F1 with 95% confidence intervals (CIs).</p>

Saved in:

Bibliographic Details
Main Author:	Koena Ronny Mabokela (21492394) (author)
Other Authors:	Mpho Primus (21492397) (author), Turgay Celik (14378019) (author)
Published:	2025
Subjects:	Ecology Sociology Immunology Science Policy Biological Sciences not elsewhere classified population without access leverage existing plms twitter sentiment dataset african languages poses advancing sentiment analysis trained language models div >< p related languages achieved existing multilingual plms sentiment analysis related languages labelled dataset multilingual plms multilingual pre developing plms resource languages native languages world ’ work expands tswana ), task adaptation specific task significant portion setswana </ sesotho </ sepedi </ safrisenti </ promising approach paper explores large data isizulu </ isixhosa </ gap leaves findings demonstrate avoiding training alternative solution 63 %.
Tags:	Add Tag No Tags, Be the first to tag this record!

Cannot write session to /tmp/vufind_sessions/sess_drkfsd55h1nrlig5o3kgqhv7il