Learning Human Detected Differences in Directed Acyclic Graphs

TL;DR

This paper introduces a machine learning approach to learn and visualize the differences in directed acyclic graphs as perceived by humans, addressing the gap between human perception and mathematical similarity measures.

Contribution

It presents a novel dataset, data augmentation algorithm, and model to capture human-detected graph differences, improving visual comparison tools.

Findings

Developed a dataset of human-labeled graph differences

Created a data augmentation algorithm for training

Achieved a model that predicts human-perceived differences

Abstract

Prior research has shown that human perception of similarity differs from mathematical measures in visual comparison tasks, including those involving directed acyclic graphs. This divergence can lead to missed differences and skepticism about algorithmic results. To address this, we aim to learn the structural differences humans detect in graphs visually. We want to visualize these human-detected differences alongside actual changes, enhancing credibility and aiding users in spotting overlooked differences. Our approach aligns with recent research in machine learning capturing human behavior. We provide a data augmentation algorithm, a dataset, and a machine learning model to support this task. This work fills a gap in learning differences in directed acyclic graphs and contributes to better comparative visualizations.