Towards Explainable and Robust NLP: Neutrosophic Probability Augmentation in Text Classification

The rapid growth of textual data necessitates advanced text classification models. However, traditional methods struggle with ambiguity and uncertainty in natural language, reducing classification reliability. To address this, we integrate neutrosophic logic, which explicitly models truth, indeterminacy, and falsity, into a DistilBERT-based text classification framework. Additionally, we employ data augmentation using synonym replacement to enhance generalization. Our approach is evaluated on the AG News dataset, classifying articles into four categories: World, Sports, Business, and Science/Technology. By incorporating neutrosophic attributes, the proposed framework assesses text quality, mitigates uncertainty, and improves robustness against ambiguous inputs. Experimental results demonstrate an accuracy of 94.10%, an F1-score of 0.9410, a precision of 0.9411, a recall of 0.9410, and a neutrosophic loss of 0.5130. Furthermore, the model achieves an average accuracy of 98.68% across five-fold cross-validation, with F1-scores consistently exceeding 0.96 and an inference rate of approximately 223 samples per second. This study highlights the effectiveness of uncertainty modeling and data augmentation in enhancing text classification performance.