Selen writeup of Brin 1999

From Cohen Courses
Jump to navigationJump to search

This is a review of Brin_1999_extracting_patterns_and_relations_from_the_world_wide_web by user:Selen


This is a very early paper of bootstrapping, they start with 5 author title pairs and they search the web looking for the occurences and extract relations from the results. Using the relations they find new occurences.

My crisitism is:

-- The usage of regular expressions: matching authors names with the proposed regrex will result in only english names

-- The results are sci-fi they could have expanded their seed list

-- There is still manual inference

--It is slow