DataNarrative: Automated Data-Driven Storytelling with Visualizations and Texts

TL;DR

This paper introduces DataNarrative, a framework using multiple LLM agents to automate the creation of data-driven stories with visualizations and text, addressing the complexity of human-like storytelling.

Contribution

It presents a novel multiagent framework for automated data story generation and a benchmark dataset of 1,449 stories from various sources.

Findings

Multiagent framework outperforms non-agentic models in evaluations.

The approach demonstrates the potential for automated, coherent data storytelling.

Challenges in generating fully coherent and comprehensive data stories remain.

Abstract

Data-driven storytelling is a powerful method for conveying insights by combining narrative techniques with visualizations and text. These stories integrate visual aids, such as highlighted bars and lines in charts, along with textual annotations explaining insights. However, creating such stories requires a deep understanding of the data and meticulous narrative planning, often necessitating human intervention, which can be time-consuming and mentally taxing. While Large Language Models (LLMs) excel in various NLP tasks, their ability to generate coherent and comprehensive data stories remains underexplored. In this work, we introduce a novel task for data story generation and a benchmark containing 1,449 stories from diverse sources. To address the challenges of crafting coherent data stories, we propose a multiagent framework employing two LLM agents designed to replicate the human storytelling process: one for understanding and describing the data (Reflection), generating the outline, and narration, and another for verification at each intermediary step. While our agentic framework generally outperforms non-agentic counterparts in both model-based and human evaluations, the results also reveal unique challenges in data story generation.