SHAP values for Explaining CNN-based Text Classification Models

TL;DR

This paper introduces a methodology for computing SHAP values to explain CNN-based text classification models, addressing interpretability challenges in NLP with high-dimensional and structured data, demonstrated on sentiment analysis.

Contribution

It develops a novel approach to compute local and global SHAP scores for CNN models in NLP, enhancing interpretability in complex text classification tasks.

Findings

Effective local explanations for CNN text classifiers.

Global feature importance scores for NLP models.

Application to sentiment analysis demonstrates practical utility.

Abstract

Deep neural networks are increasingly used in natural language processing (NLP) models. However, the need to interpret and explain the results from complex algorithms are limiting their widespread adoption in regulated industries such as banking. There has been recent work on interpretability of machine learning algorithms with structured data. But there are only limited techniques for NLP applications where the problem is more challenging due to the size of the vocabulary, high-dimensional nature, and the need to consider textual coherence and language structure. This paper develops a methodology to compute SHAP values for local explainability of CNN-based text classification models. The approach is also extended to compute global scores to assess the importance of features. The results are illustrated on sentiment analysis of Amazon Electronic Review data.