Recovering semantic relations from web pages based on visual cues

  • Authors:
  • Peifeng Xiang;Yuanchun Shi

  • Affiliations:
  • Tsinghua University, Beijing, China;Tsinghua University, Beijing, China

  • Venue:
  • Proceedings of the 11th international conference on Intelligent user interfaces
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recovering semantic relations between different parts of web pages are of great importance for multi-platform web interface development, as they make it possible to re-distribute interaction objects and change the structure of interfaces while preserving the semantics of the UI. Important semantic relations include topic, order, hierarchy, etc. This paper presents a visual cues based approach, which is tag-tree structure independent, to automatically detect such kind of semantic relations in web pages. Comparing with other existing techniques, such as DOM-based methods, this approach mostly depends on interfaces' perceptible visual information that is more reliable. The preliminary evaluation on complex web sites shows promising results. We believe further exploration is worth taken.