Author: paul.rosen

Topox: A Suite Of Python Packages For Machine Learning On Topological Domains

We introduce TopoX, a Python software suite that provides reliable and user-friendly building blocks for computing and machine learning on topological domains that extend graphs: hypergraphs, simplicial, cellular, path and combinatorial complexes. TopoX consists of three packages: TopoNetX facilitates constructing and computing on these domains, including working with nodes, edges and higher-order cells; TopoEmbedX provides methods to embed topological domains into vector spaces, akin to popular graph-based embedding algorithms such as node2vec; TopoModelX is built on top of PyTorch and offers a comprehensive toolbox of higher-order message passing functions for neural networks on topological domains.

Continue reading

Enhancing Student Feedback Using Predictive Models In Visual Literacy Courses

This study looks at employing Naive Bayes modeling to analyze peer review data from an undergraduate visual literacy course over a five-year period (2017-2022). A key finding is the emphasis on the ‘lie factor’ in students’ comments when using the visual peer review rubric, highlighting areas for potential course content adaptation and instructional refinement. Our findings suggest that predictive modeling, as a tool to assess student comments, can provide novel insights into visual peer review. This could lead to impactful changes in course content, project modifications, and rubric enhancements, ultimately benefiting student learning and engagement in visual literacy courses.

Continue reading

CLAMS: Cluster Ambiguity Measure for Estimating Perceptual Variability in Visual Clustering

We introduce CLAMS, a data-driven visual quality measure for automatically predicting cluster ambiguity in monochrome scatterplots. We first conduct a qualitative study to identify key factors that affect the visual separation of clusters (e.g., proximity or size difference between clusters). Based on the study findings, we deploy a regression module that estimates the human-judged separability of two clusters. Then, CLAMS predicts cluster ambiguity by analyzing the aggregated results of all pairwise separability between clusters that are generated by the module. CLAMS outperforms widely-used clustering techniques in predicting ground truth cluster ambiguity.

Continue reading

A Comparative Study Of The Perceptual Sensitivity Of Topological Visualizations To Feature Variations

This paper evaluates the sensitivity of topology-based isocontour, Reeb graph, and persistence diagram visualizations compared to a reference color map visualization for synthetically generated scalar fields on 2-manifold triangular meshes embedded in 3D. In particular, we built and ran a human-subject study that evaluated the perception of data features characterized by Gaussian signals and measured how effectively each visualization technique portrays variations of data features arising from the position and amplitude variation of a mixture of Gaussians.

Continue reading

Exploring Annotation Strategies In Professional Visualizations: Insights From Prominent US News Portals

Annotations play a vital role in visualizations, providing valuable insights and focusing attention on critical visual elements. This study analyzes a curated corpus of 72 professionally designed static charts with annotations from prominent US news portals including The New York Times, The Economists, The Wall Street Journal, and The Washington Post. The analysis reveals common patterns in annotation strategies used by professionals.

Continue reading

Homology-Preserving Multi-Scale Graph Skeletonization Using Mapper On Graphs

We apply the mapper construction—a popular tool in topological data analysis—to graph visualization, which provides a strong theoretical basis for summarizing the data while preserving their core structures. We develop a variation of the mapper construction targeting weighted, undirected graphs, called Mapper on Graphs, which generates homology-preserving skeletons of graphs. We further show how the adjustment of a single parameter enables multi-scale skeletonization of the input graph. Finally, we provide a software tool that enables interactive explorations of such skeletons and demonstrates the effectiveness of our method for synthetic and real-world data.

Continue reading

Topological Deep Learning: Going Beyond Graph Data

In this paper, we present a unifying deep learning framework built upon a richer data structure that includes widely adopted topological domains. Specifically, we first introduce combinatorial complexes, a novel type of topological domain. Second, we develop a general class of message-passing combinatorial complex neural networks (CCNNs), focusing primarily on attention-based CCNNs. Third, we evaluate the performance of CCNNs on tasks related to mesh shape analysis and graph learning.

Continue reading

Untangling Force-Directed Layouts Using Persistent Homology

In this paper, we use the principles of persistent homology to untangle force-directed layouts thus mitigating these issues. First, we devise a new method to use 0-dimensional persistent homology to efficiently generate an initial graph layout, resulting in faster convergence and better quality graph layouts. Second, we provide an efficient algorithm for 1-dimensional persistent homology features and provide users the ability to interact with the 1-dimensional features by highlighting them and adding cycle-emphasizing forces to the layout.

Continue reading

A Qualitative Evaluation And Taxonomy Of Student Annotations On Bar Charts

Annotations have become an essential part of visualizations, primarily when externalizing data or engaging in collaborative analysis. Therefore, it is crucial to understand how people annotate visualizations. This two-phase study used individual and group settings to investigate how visualization students annotate bar charts when asked to answer high-level questions about the data in the charts. The resulting annotations were coded and summarized into a taxonomy with several interesting findings.

Continue reading

Automatic Scatterplot Design Optimization For Clustering Identification

In this paper, we propose an automatic tool to optimize the design factors of scatterplots to reveal the most salient cluster structure. Our approach leverages the merge tree data structure to identify the clusters and optimize the choice of subsampling algorithm, sampling rate, marker size, and marker opacity used to generate a scatterplot image.

Continue reading