Using structured tokens to identify webpages for data extraction

  • Authors:

  • Ling Lin;Lizhu Zhou;Qi Guo;Gang Li

  • Affiliations:

  • Tsinghua University, Beijing, PRC;Tsinghua University, Beijing, PRC;Tsinghua University, Beijing, PRC;Tsinghua University, Beijing, PRC

  • Venue:
  • APWeb/WAIM'07 Proceedings of the joint 9th Asia-Pacific web and 8th international conference on web-age information management conference on Advances in data and web management
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

As the web grows, more and more data has become available from webpages, such as the product items from the back-end databases. To provide efficient access to the data objects contained in these pages, data extraction plays an important role. However, identifying the suitable webpages to feed the data extraction is a pre-requisite and non-trivial task. As a result, there is an increasing need for methods that can automatically identify the target pages from unknown websites. In this paper, we solve the problem by exploiting the structured-token features of the webpage content, and applying decision tree based classification algorithm to induce the structure information. Furthermore, a preliminary recognition of data-object is acquired to efficiently initiate the subsequential data extraction. We experiment our approach on the real-world data, and achieve promising results.