# Rating Sentiment Analysis Systems for Bias through a Causal Lens

arXiv: https://arxiv.org/abs/2302.02038 · HTML: https://mlsanity.com/paper/2302.02038/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 47 references that are in the map.
- First posted on arXiv 4 Feb 2023.

## Builds on (most cited first, 20 of 20)

- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [Examining Gender and Race Bias in Two Hundred Sentiment Analysis Systems](https://mlsanity.com/paper/1805.04508/) (2018)
- [On instabilities of deep learning in image reconstruction - Does AI come at a cost?](https://mlsanity.com/paper/1902.05300/) (2019)
- [Racial Bias in Hate Speech and Abusive Language Detection Datasets](https://mlsanity.com/paper/1905.12516/) (2019)
- [Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyond](https://mlsanity.com/paper/2109.00725/) (2021)
- [Reducing Gender Bias in Abusive Language Detection](https://mlsanity.com/paper/1808.07231/) (2018)
- [Assessing Gender Bias in Machine Translation -- A Case Study with Google Translate](https://mlsanity.com/paper/1809.02208/) (2018)
- [Challenges in Automated Debiasing for Toxic Language Detection](https://mlsanity.com/paper/2102.00086/) (2021)
- [Equalizing Gender Biases in Neural Machine Translation with Word Embeddings Techniques](https://mlsanity.com/paper/1901.03116/) (2019)
- [Generative Interventions for Causal Learning](https://mlsanity.com/paper/2012.12265/) (2020)
- [Causal Fairness Analysis](https://mlsanity.com/paper/2207.11385/) (2022)
- [On the Legal Compatibility of Fairness Definitions](https://mlsanity.com/paper/1912.00761/) (2019)
- [Deconfounded Causal Collaborative Filtering](https://mlsanity.com/paper/2110.07122/) (2021)
- [Unicorn: Reasoning about Configurable System Performance through the lens of Causality](https://mlsanity.com/paper/2201.08413/) (2022)
- [Bias in Data-driven AI Systems -- An Introductory Survey](https://mlsanity.com/paper/2001.09762/) (2020)
- [Towards Composable Bias Rating of AI Services](https://mlsanity.com/paper/1808.00089/) (2018)
- [Transfer Learning for Performance Modeling of Configurable Systems: A Causal Analysis](https://mlsanity.com/paper/1902.10119/) (2019)

## Built on by (most cited first, 6 of 6)

- [The Effect of Human v/s Synthetic Test Data and Round-tripping on Assessment of Sentiment Analysis Systems for Bias](https://mlsanity.com/paper/2401.12985/) (2024)
- [Rating Multi-Modal Time-Series Forecasting Models (MM-TSFM) for Robustness Through a Causal Lens](https://mlsanity.com/paper/2406.12908/) (2024)
- [Holistic Explainable AI (H-XAI): Extending Transparency Beyond Developers in AI-Driven Decision Making](https://mlsanity.com/paper/2508.05792/) (2025)
- [On Identifying Why and When Foundation Models Perform Well on Time-Series Forecasting Using Automated Explanations and Rating](https://mlsanity.com/paper/2508.20437/) (2025)
- [Trust and ethical considerations in a multi-modal, explainable AI-driven chatbot tutoring system: The case of collaboratively solving Rubik's Cube](https://mlsanity.com/paper/2402.01760/) (2024)
- [Creating a Causally Grounded Rating Method for Assessing the Robustness of AI Models for Time-Series Forecasting](https://mlsanity.com/paper/2502.12226/) (2025)

## Abstract (arXiv)

Sentiment Analysis Systems (SASs) are data-driven Artificial Intelligence (AI) systems that, given a piece of text, assign one or more numbers conveying the polarity and emotional intensity expressed in the input. Like other automatic machine learning systems, they have also been known to exhibit model uncertainty where a (small) change in the input leads to drastic swings in the output. This can be especially problematic when inputs are related to protected features like gender or race since such behavior can be perceived as a lack of fairness, i.e., bias. We introduce a novel method to assess and rate SASs where inputs are perturbed in a controlled causal setting to test if the output sentiment is sensitive to protected variables even when other components of the textual input, e.g., chosen emotion words, are fixed. We then use the result to assign labels (ratings) at fine-grained and overall levels to convey the robustness of the SAS to input changes. The ratings serve as a principled basis to compare SASs and choose among them based on behavior. It benefits all users, especially developers who reuse off-the-shelf SASs to build larger AI systems but do not have access to their code or training data to compare.