WWW2005 Poster

1
The Chinese University of Hong Kong
Web Page Classification with Heterogeneous Data
Fusion
Zenglin Xu, Irwin King and Michael R.
Lyu Department of Computer Science and
Engineering The Chinese University of Hong
Kong zlxu, king, lyu_at_cse.cuhk.edu.hk
Motivations
1
2
Contributions

• For web page classification, there are many
  available data
• sources, such as the text, the title, the meta
  data, the anchor
• text, etc.
• Simply putting them together would not greatly
  enhance the
• classification performance.
• Different dimensions and types of data sources
  can be
• represented into a common format of kernel
  matrix.
• A kernel learning approach is thus proposed to
  integrate
• multiple data sources

• A systematic way of integrating multiple
• data sources.
• Better classification accuracy.

Architacture Model
3

• 1. Feature Extraction.
• 2. Similarity Representation. Each data source
  is represented as a kernel matrix (Ki)
• 3. Similarity Combination.
• 4. Classification.
• Substitute K into the dual SVM
• We have the following QCQP problem

Experiment results
4

• Dataset DMOZ
• AT Anchor Text
• LT Link Text
• MT Meta Data
• TI Title
• PT Plain Text
• UW Universally Weighted sources
• KC sources by Kernel Combination
• Mi -F1 Micro-F1
• Ma-F1 Macro-F1

WWW 2007, May 812, 2007, Banff, Alberta, Canada.
The Chinese University of Hong Kong