Philgoo Han writeup of Wang and Cohen, 2007

This is a review of Wang_2007_language_independent_set_expansion_of_named_entities_using_the_web by user:Ironfoot.

• SEAL (language independent set expander)
  • Fetcher
    • Fetch google queries with given set of (small) seeds
  • Extractor
    • Page dependent
    • Logest common sequence character on prefix and suffix of found seed.
  • Ranker
    • Filter out noisy entities
    • Graph of seeds, web pages, wrappers and extracted entities
    • Find similarity between nodes by graph walk
  • Result
    • Roughly twice better than Google Sets
    • Handle multiple language