Divisi: Interactive Search and Visualization for Scalable Exploratory Subgroup Analysis

2025-02-14

AI Summary

Identifying manually defined subgroups in high-dimensional data is challenging, and existing methods lack support for exploratory subgroup discovery. To address this, we propose a novel paradigm—exploratory subgroup analysis—and introduce the “Subgroup Map,” a visual analytics framework enabling dynamic reordering, interactive refinement, and quantitative analysis of subgroup coverage relationships. We further design a lightweight, approximate subgroup discovery algorithm, seamlessly integrated into Jupyter notebooks. Our approach balances computational efficiency with interpretability, supporting automatic subgroup detection, multi-dimensional evaluation (e.g., statistical significance, coverage, divergence), and iterative filtering. In an empirical study with 13 data scientists, our method significantly improved both the speed and depth of subgroup discovery, successfully uncovering unexpected feature interactions and fine-grained distributional biases. The Subgroup Map provides a new analytical tool for understanding data heterogeneity and model behavior.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Dimensionality Reduction/Feature Selection
• Humans and AI: Planning and Decision Support for Human-Machine Teams

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Web data visualization
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Analyzing data subgroups is a common data science task to build intuition about a dataset and identify areas to improve model performance. However, subgroup analysis is prohibitively difficult in datasets with many features, and existing tools limit unexpected discoveries by relying on user-defined or static subgroups. We propose exploratory subgroup analysis as a set of tasks in which practitioners discover, evaluate, and curate interesting subgroups to build understanding about datasets and models. To support these tasks we introduce Divisi, an interactive notebook-based tool underpinned by a fast approximate subgroup discovery algorithm. Divisi's interface allows data scientists to interactively re-rank and refine subgroups and to visualize their overlap and coverage in the novel Subgroup Map. Through a think-aloud study with 13 practitioners, we find that Divisi can help uncover surprising patterns in data features and their interactions, and that it encourages more thorough exploration of subtypes in complex data.

Problem

• Interactive tool for scalable subgroup analysis
• Discover, evaluate, and curate interesting subgroups
• Visualize subgroup overlap and coverage effectively

Innovation

• Interactive notebook-based tool
• Fast approximate subgroup discovery
• Novel Subgroup Map visualization