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

Solving graph data issues using a layered architecture approach with applications to web spam detection.

Franco Scarselli1, Ah Chung Tsoi, Markus Hagenbuchner

  • 1University of Siena, Siena, Italy.

Neural Networks : the Official Journal of the International Neural Network Society
|August 27, 2013
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

Leveraging molecular graphs for natural product classification.

Computational and structural biotechnology journal·2026
Same author

Point cloud dosimetry framework for preclinical microbeam radiation therapy.

Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics (AIFB)·2025
Same author

Interoperable Traceability in Agrifood Supply Chains: Enhancing Transport Systems Through IoT Sensor Data, Blockchain, and DataSpace.

Sensors (Basel, Switzerland)·2025
Same author

Comparison of Deep Learning Models for fast and accurate dose map prediction in Microbeam Radiation Therapy.

Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics (AIFB)·2025
Same author

Dual-Branch Multi-Dimensional Attention Mechanism for Joint Facial Expression Detection and Classification.

Sensors (Basel, Switzerland)·2025
Same author

An Innovative Neighbor Attention Mechanism Based on Coordinates for the Recognition of Facial Expressions.

Sensors (Basel, Switzerland)·2024

This study combines unsupervised and supervised learning for graph data processing. The novel cascade architecture improves graph neural network performance, effectively detecting web spam and achieving state-of-the-art results.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Data Science

Background:

  • Graph data processing is crucial for many applications.
  • Graph Neural Networks (GNNs) are powerful but face challenges like long-term dependencies.
  • Unsupervised learning methods can potentially mitigate these issues.

Purpose of the Study:

  • To propose a novel cascade architecture combining probabilistic mapping graph self-organizing maps (PMG-SOM) and GNNs.
  • To address and limit the long-term dependency problem in GNN training.
  • To enhance the overall performance of graph-based learning tasks.

Main Methods:

  • A cascade architecture integrating PMG-SOM (unsupervised) with GNN (supervised).
  • Utilizing PMG-SOM to preprocess or inform the GNN input.
Keywords:
Graph neural networkLayered architectureProbability Mapping GraphSOMWeb spam detection

Related Experiment Videos

  • Evaluating the combined approach on a web spam detection benchmark dataset.

    • Main Results:

    • The proposed cascade architecture successfully limits long-term dependencies in GNNs.
    • Demonstrated significant performance improvements compared to standalone GNNs.
    • Achieved state-of-the-art results on the web spam detection task, comparable to diverse existing methods.

    Conclusions:

    • The combined PMG-SOM and GNN approach offers a robust solution for graph data processing.
    • This method shows broad applicability to any domain representable as a graph.
    • The architecture effectively enhances GNN performance and tackles key training challenges.