![[2602.16120] Feature-based morphological analysis of shape graph data](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.16120] Feature-based morphological analysis of shape graph data
Summary
This paper presents a computational pipeline for analyzing shape graph datasets, focusing on geometric and topological features to enhance tasks like clustering and classification.
Why It Matters
Understanding the geometric and topological variations in shape graphs is crucial for applications in urban planning, neuroscience, and imaging. This research offers a novel approach to analyze complex datasets, potentially improving decision-making in various fields.
Key Takeaways
- Introduces a new computational pipeline for shape graph analysis.
- Focuses on extracting topological, geometric, and directional features.
- Demonstrates effectiveness on real-world datasets like urban networks and neuronal traces.
- Benchmarks against existing methods to validate the proposed approach.
- Aims to enhance clustering and classification tasks in complex datasets.
Computer Science > Machine Learning arXiv:2602.16120 (cs) [Submitted on 18 Feb 2026] Title:Feature-based morphological analysis of shape graph data Authors:Murad Hossen, Demetrio Labate, Nicolas Charon View a PDF of the paper titled Feature-based morphological analysis of shape graph data, by Murad Hossen and 2 other authors View PDF HTML (experimental) Abstract:This paper introduces and demonstrates a computational pipeline for the statistical analysis of shape graph datasets, namely geometric networks embedded in 2D or 3D spaces. Unlike traditional abstract graphs, our purpose is not only to retrieve and distinguish variations in the connectivity structure of the data but also geometric differences of the network branches. Our proposed approach relies on the extraction of a specifically curated and explicit set of topological, geometric and directional features, designed to satisfy key invariance properties. We leverage the resulting feature representation for tasks such as group comparison, clustering and classification on cohorts of shape graphs. The effectiveness of this representation is evaluated on several real-world datasets including urban road/street networks, neuronal traces and astrocyte imaging. These results are benchmarked against several alternative methods, both feature-based and not. Subjects: Machine Learning (cs.LG); Applications (stat.AP); Machine Learning (stat.ML) MSC classes: 62R30, 62P10 Cite as: arXiv:2602.16120 [cs.LG] (or arXiv:2602.16120v1 [...
