Data storytelling improves ai decision explanations for non-experts
Data storytelling meets interpretable machine learning: Decoding AI decisions for non-experts without revealing sensitive data and model details
Artificial IntelligenceComputation and LanguageHuman-Computer InteractionMachine Learning
Summary
AI systems often make decisions that are accurate but hard for regular people to understand. The authors created a way to tell simple, story-like explanations about how AI makes decisions without showing sensitive data or complicated details. They tested this on a well-known dataset and found most people understood these explanations better than traditional charts. This approach helps more people trust and use AI decisions by making them easier to follow.
What this means in practice
- •For business analysts: Create easy-to-understand AI decision explanations that protect sensitive data during client presentations and reporting.
- •For financial compliance teams: Use narrative explanations to interpret AI risk assessments without exposing proprietary model details or private data.
Tested on one dataset.
Authors
Lemen Chao, Zixuan Yang, Anran Fang, Mingran Sun, Ming Lei
Abstract
AI-driven automated decision-making requires both predictive performance and interpretability. Recent advances in interpretable machine learning (IML) provide tools for explaining model predictions, but the technical complexity of these explanations may hinder accessibility to non-experts. To address this challenge, this study integrates data storytelling with IML to enhance the explainability of AI-generated decisions for a broader audience. Following the design science research (DSR) paradigm, this study proposes a formal definition of data storytelling in IML, introduces the DIST Pyramid to align data storytelling with IML, and presents the I-P-O Model to describe their interactions. It further develops an architecture to explain AI decisions through distinct "What-if" and "Why-not" event-generation processes. The architecture also employs data desensitization to protect sensitive input data. To validate the approach, a case study is conducted with the Boston Housing dataset, using SHapley Additive exPlanations (SHAP) values and large language models (LLMs) to generate data stories with And-But-Therefore (ABT) structures. An empirical evaluation shows that 76.4% and 74.3% of respondents rated the "What-if" and "Why-not" data stories as more comprehensible, with significantly higher accessibility scores than traditional SHAP visualizations. The paper concludes with the presentation of a narrative interpretation framework that integrates IML and data storytelling, thereby expanding the research scope as well as the practical applicability of AI decision-making.