WeStcoin: Weakly-Supervised Contextualized Text Classification with Imbalance and Noisy Labels

The joint problem of imbalance samples and noisy labels challenges the current text classifiers in real-world applications. Existing approaches are mostly devoted to handling either former or latter while fail to manage the fused issue. This paper introduces a novel weakly-supervised framework, dubbed WeStcoin, to take into account the sensitivity cost on misclassifications between classes and seek seed words towards noisy-label corrections. After BERT that creates a contextualized corpus, WeSt-coin learns a predicted label vector from the contextualized samples and meanwhile calculates a pseudo probability vector from seed words, and then projects the concatenated representation into an output space, followed by multiplying by a cost-sensitive matrix. WeStcoin is ultimately trained to decrease the residual between the model outputs and the noisy labels, where seed words are also updated in an iterative manner. Extensive experiments and ablation studies on two public text datasets demonstrate that the proposed model outperforms the state-of-the-art model in the text classification with imbalance samples and noisy labels. Codes are made available at https://github.com/ypzhaang.