Update README.md

README.md

@@ -24,7 +24,7 @@ It achieves the following results on the evaluation set:
 
 ## Model description
 
-Data was labelled according to review stars. If stars >3, review was ranked positive. Otherwise, it is labelled as negative. We have tried 4 stars instead of 3 as app developers would deem any negativity in reviews as negative as a whole, but accuracy dropped. Further investigations will need to be run.
+Data was labelled according to review stars. If stars >3, review was ranked positive. Otherwise, it is labelled as negative. We have tried 4 stars instead of 3 as app developers would deem any negativity in reviews as negative as a whole, but accuracy dropped. (As of december 2024, I realised that the accuracy dropped as the 3 stars data high accuracy could be due to overfitting AND imbalanced data. I know better now.) Further investigations will need to be run.
 Above 4 stars positive: https://huggingface.co/ajiayi/finetuning-sentiment-model-bank_reviews-otherbank-4insteadof3
 All data (OCBC,POSB,GXS): https://huggingface.co/ajiayi/finetuning-sentiment-model-bank_reviews