Hierarchical Interpretation of Neural Text Classification

TL;DR

This paper introduces Hint, a hierarchical neural text classifier that explains predictions through label-associated topics, providing more faithful and human-understandable interpretations while maintaining competitive classification performance.

Contribution

It proposes a novel hierarchical interpretability method that explains neural text classification via topics instead of words, improving faithfulness and interpretability.

Findings

Achieves classification accuracy comparable to state-of-the-art models.

Generates more faithful and understandable explanations.

Effective on review and news datasets.

Abstract

Recent years have witnessed increasing interests in developing interpretable models in Natural Language Processing (NLP). Most existing models aim at identifying input features such as words or phrases important for model predictions. Neural models developed in NLP however often compose word semantics in a hierarchical manner and text classification requires hierarchical modelling to aggregate local information in order to deal with topic and label shifts more effectively. As such, interpretation by words or phrases only cannot faithfully explain model decisions in text classification. This paper proposes a novel Hierarchical INTerpretable neural text classifier, called Hint, which can automatically generate explanations of model predictions in the form of label-associated topics in a hierarchical manner. Model interpretation is no longer at the word level, but built on topics as the basic semantic unit. Experimental results on both review datasets and news datasets show that our proposed approach achieves text classification results on par with existing state-of-the-art text classifiers, and generates interpretations more faithful to model predictions and better understood by humans than other interpretable neural text classifiers.