Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Learning shape-classes using a mixture of tree-unions.

Andrea Torsello1, Edwin R Hancock

  • 1Dipartimento di Informatica, Università Ca' Foscari di Venezia, via Torino 155, 30172 Venezia Mestre, Italy. torsello@dsi.unive.it

IEEE Transactions on Pattern Analysis and Machine Intelligence
|May 27, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Learning From Human Attention for Attribute-Assisted Visual Recognition.

IEEE transactions on pattern analysis and machine intelligence·2024
Same author

The Ihara zeta function as a partition function for network structure characterisation.

Scientific reports·2024
Same author

Graph Kernel Neural Networks.

IEEE transactions on neural networks and learning systems·2024
Same author

Regionwise Generative Adversarial Image Inpainting for Large Missing Areas.

IEEE transactions on cybernetics·2022
Same author

Deep Demosaicing for Polarimetric Filter Array Cameras.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2022
Same author

Two-Level Graph Neural Network.

IEEE transactions on neural networks and learning systems·2022
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces a novel tree-clustering method using mixture models and minimum description length. It effectively infers node correspondences for classifying 2D shapes represented by shock graphs.

Area of Science:

  • Machine Learning
  • Computational Biology
  • Pattern Recognition

Background:

  • Tree-structured data is prevalent in various scientific domains.
  • Clustering tree data is challenging due to unknown node correspondences.
  • Existing methods often struggle with inferring relationships within tree structures.

Purpose of the Study:

  • To develop a robust tree-clustering algorithm that handles unknown node correspondences.
  • To apply a mixture model approach for grouping similar tree structures.
  • To leverage the minimum description length principle for model fitting.

Main Methods:

  • Formulating tree-clustering as fitting a mixture of tree unions.
  • Assuming a Bernoulli distribution for observed tree nodes within clusters.

Related Experiment Videos

  • Employing maximum-likelihood estimation for Bernoulli parameters.
  • Utilizing a minimum description length (MDL) criterion to optimize tree unions and mixing proportions.
  • Inferring node correspondences by minimizing edit distance.

    • Main Results:

    • The proposed method successfully clusters tree data even when node correspondences are unknown.
    • Edit distance is shown to be intrinsically linked to the description length criterion.
    • The algorithm is applicable to both unweighted and weighted trees.
    • Demonstrated effectiveness in classifying 2D shapes using shock graph representations.

    Conclusions:

    • The developed tree-clustering method offers a powerful approach for analyzing complex tree-structured data.
    • The minimum description length principle provides an effective framework for model fitting and inference in tree clustering.
    • The algorithm shows promise for applications in shape classification and other areas involving structural data analysis.