Wednesday, February 1, 2012

How Google uses thesaurus in search - an interesting example


Look at this a search query:
https://www.google.com/search?q=extroverted+personality+essay
(note the word "essay")

Check out the first (as of Feb.1, 2012) result. The word "paper" is bold:

Research Paper on Introvert vs. Extrovert Personalities - College ...
www.academon.com › Psychology › Behaviorism
Research Paper on Introvert vs. Extrovert Personalities - Library of College Term Papers, Research Papers, Essays and Book Reports.
  1. Google thinks that "paper" and "essay" are synonyms.
  2. Google knows that "paper" means "research paper" or "term paper" and it's not about printers and copy machines or parchment / waxed paper.
  3. Google does not highlight the plural form, "papers". It will if you search for "essays" in plural.
So far, it's all interesting observations, but here is what I have difficulty explaining:
If you look at the page itself, you will see a tons of "paper" in the content and just a couple of "essay". BUT: if you replace "essay" with "paper" in the search query, this page drops down from #1 to #4.
So, Google prefers a synonym to the exact match in this case!

What do you think? Please comment.