# Multi-resolution Interpretation and Diagnostics Tool for Natural Language Classifiers

arXiv: https://arxiv.org/abs/2303.03542 · HTML: https://mlsanity.com/paper/2303.03542/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 31 references that are in the map.
- First posted on arXiv 6 Mar 2023.

## Builds on (most cited first, 13 of 13)

- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [Model-Agnostic Interpretability of Machine Learning](https://mlsanity.com/paper/1606.05386/) (2016)
- [Post-hoc Interpretability for Neural NLP: A Survey](https://mlsanity.com/paper/2108.04840/) (2021)
- [iNNvestigate neural networks!](https://mlsanity.com/paper/1808.04260/) (2018)
- ["What is Relevant in a Text Document?": An Interpretable Machine Learning Approach](https://mlsanity.com/paper/1612.07843/) (2016)
- [Explanation-Based Human Debugging of NLP Models: A Survey](https://mlsanity.com/paper/2104.15135/) (2021)
- [Interpreting Black Box Predictions using Fisher Kernels](https://mlsanity.com/paper/1810.10118/) (2018)
- [Locally Interpretable Models and Effects based on Supervised Partitioning (LIME-SUP)](https://mlsanity.com/paper/1806.00663/) (2018)
- [Practical and Ethical Considerations in the Effective use of Emotion and Sentiment Lexicons](https://mlsanity.com/paper/2011.03492/) (2020)
- [Model Explainability in Deep Learning Based Natural Language Processing](https://mlsanity.com/paper/2106.07410/) (2021)
- [Model Interpretation: A Unified Derivative-based Framework for Nonparametric Regression and Supervised Machine Learning](https://mlsanity.com/paper/1808.07216/) (2018)
- [Text Length Adaptation in Sentiment Classification](https://mlsanity.com/paper/1909.08306/) (2019)
- [Model-Agnostic Explanations using Minimal Forcing Subsets](https://mlsanity.com/paper/2011.00639/) (2020)

## Abstract (arXiv)

Developing explainability methods for Natural Language Processing (NLP) models is a challenging task, for two main reasons. First, the high dimensionality of the data (large number of tokens) results in low coverage and in turn small contributions for the top tokens, compared to the overall model performance. Second, owing to their textual nature, the input variables, after appropriate transformations, are effectively binary (presence or absence of a token in an observation), making the input-output relationship difficult to understand. Common NLP interpretation techniques do not have flexibility in resolution, because they usually operate at word-level and provide fully local (message level) or fully global (over all messages) summaries. The goal of this paper is to create more flexible model explainability summaries by segments of observation or clusters of words that are semantically related to each other. In addition, we introduce a root cause analysis method for NLP models, by analyzing representative False Positive and False Negative examples from different segments. At the end, we illustrate, using a Yelp review data set with three segments (Restaurant, Hotel, and Beauty), that exploiting group/cluster structures in words and/or messages can aid in the interpretation of decisions made by NLP models and can be utilized to assess the model's sensitivity or bias towards gender, syntax, and word meanings.