Data Formulator 2: Iterative Creation of Data Visualizations, with AI Transforming Data Along the Way

2024-08-28

AI Summary

Current AI-augmented visualization systems rely on single-turn text prompts, limiting their ability to support the dynamic, iterative nature of exploratory data analysis. To address this, we propose VisIterate—a collaborative, iterative framework that synergistically integrates GUI-based and natural-language interactions. VisIterate introduces the first mechanism for joint evolution of visualization design and data transformation, enabled by three core innovations: (1) visualization state versioning, (2) iterative intent modeling, and (3) history-aware navigation and reuse—thereby transcending the constraints of conventional single-prompt paradigms. The system automatically executes semantic-driven data transformations while allowing users to refine visual encodings interactively via the GUI. An empirical user study (n=8) demonstrates that participants rapidly develop personalized iterative strategies and achieve significantly improved efficiency in complex exploratory tasks. This work establishes a scalable, iterative paradigm for AI-enhanced interactive visualization.

Technology Category

• Humans and AI: Intelligent User Interfaces
• Data Mining & Knowledge Management: Data Visualization & Summarization
• Computer Vision: Visual Reasoning & Symbolic Representations

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Assisted, interactive, and conversational search

Abstract

Data analysts often need to iterate between data transformations and chart designs to create rich visualizations for exploratory data analysis. Although many AI-powered systems have been introduced to reduce the effort of visualization authoring, existing systems are not well suited for iterative authoring. They typically require analysts to provide, in a single turn, a text-only prompt that fully describe a complex visualization. We introduce Data Formulator 2 (DF2 for short), an AI-powered visualization system designed to overcome this limitation. DF2 blends graphical user interfaces and natural language inputs to enable users to convey their intent more effectively, while delegating data transformation to AI. Furthermore, to support efficient iteration, DF2 lets users navigate their iteration history and reuse previous designs, eliminating the need to start from scratch each time. A user study with eight participants demonstrated that DF2 allowed participants to develop their own iteration styles to complete challenging data exploration sessions.

Problem

• Enables iterative data visualization creation.
• Integrates AI for data transformation tasks.
• Supports efficient iteration with history navigation.

Innovation

• AI-powered visualization system
• Graphical and natural language interfaces
• Iteration history navigation