Low confidence predictions analysis
How do examples where the model is correct but the confidence of classification is low look like? In this case, the minimum probability of a predicted class can be 0.33 (⅓):
If the model predicts correctly with 0.35, check those examples and see if they are really hard to identify.
If the model predicts an obviously positive remark like ‘I am so happy for the good work I have done’ correctly with probability 0.35, something is fishy.