New platform enables machine learning on rendered web pages

Researchers have developed a new platform designed to make machine learning applications on web pages more feasible. This platform renders web pages, capturing their visual and structural properties in a Resource Description Framework (RDF) format. It includes a server for rendering, a REST API for control, and SPARQL queries for data retrieval, enabling the creation of datasets suitable for machine learning. A Python client library integrates this platform into standard machine learning workflows, demonstrated by training graph neural networks on rendered web pages to identify key content elements. AI

IMPACT Enables new approaches to analyzing and extracting information from web content using machine learning.

RANK_REASON The item is an academic paper detailing a new platform and methodology for machine learning on web pages. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.IR (Information Retrieval) →

• alphaXiv
• CatalyzeX
• Connected Papers
• CORE Recommender
• DagsHub
• FitLayout
• Gotit.pub
• graph neural networks
• Hugging Face
• IArxiv Recommender
• Litmaps
• Python
• Resource Description Framework
• ScienceCast
• scite Smart Citations
• SPARQL

• paper
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →