multi-class classification using SVM

I'm working on a multi-class classification problem, there are 4 classes but the dataset is imbalanced.
SVM may not be the best choice but I'm testing different algorithms,
The results aren't that bad but there is one class that particularly has a lower accuracy

              precision    recall  f1-score   support

           0       0.73      0.85      0.79       117
           1       0.90      0.88      0.89       666
           2       0.91      0.91      0.91       767
           3       0.87      0.86      0.87       168

    accuracy                           0.89      1718
   macro avg       0.85      0.87      0.86      1718
weighted avg       0.89      0.89      0.89      1718

I have tried to adjust class_weight but the results got worse
is there a simple way where I can adjust the parameters to have higher accuracy for that class?