AI4D iCompass Social Media Sentiment Analysis for Tunisian Arabizi
$2,000 USD
Can you classify sentiment in the Tunisian Arabizi dialect?
467 data scientists enrolled, 171 on the leaderboard
20 November 2020—28 March 2021
Ends in 2 months
Why not f1 score for Evaluation metrics?
published 2 Jan 2021, 15:36

I think the best-overfitted model will be winning the competition.

As the classes are imbalanced, accuracy may not be valid metrics if the test dataset is imbalanced. What do you think?

what makes you think that the test dataset is imbalanced ?

i am totally agree with you