Extracting Schema from Semistructured Data with Weight Tag

  • Authors:

  • Jiuzhong Li;Shuo Shi

  • Affiliations:

  • Department of Computer Engineering, Guangdong Industry Technical College, Guangzhou, China 510300;Department of Computer Engineering, Guangdong Industry Technical College, Guangzhou, China 510300

  • Venue:
  • ISNN 2009 Proceedings of the 6th International Symposium on Neural Networks: Advances in Neural Networks - Part III
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper put forward the concept of OEM model with weight on its edges, developes a new approach to extracting schema from semistructured data with weight on its edges, and gives two theorems related to computing taget set of label path and suporting degree of label path. Using wideth-first and top-down traversing strategy ,the algorithm computes target set and supporting degree of every label in a label path, and decides whether the label is retained in schema model according to its magnitude of supporting degree and weight of the label .In the last, we test the validity and efficiency of the algorithm. The schema scale of the semistructured data obtained from the same OEM database in this paper is smaller than that in other paper.