AOL recently released a huge sample of search engine queries. In a highly questionable move they tied these queries to reasonably anonomous user identifiers; for example we know that user known as #724 searched for “how to install a glue down floor”, as well as “carbol tunnel” etc. He did 366 searchs between March 1st and May 5 2006.

Unsuprisingly the distribution of search generators is power-law distributed. This is a log-log chart. Each dot on the chart represents one AOL users. The vertical axis is how many searches they did; for example the highest dot, aka user 2263543, did 8695 searches. This is only the most active 20 thousand users in this data set, the least active of whom did 313 searches. The complete data set has 657 thousand users, 57 thousand of whom only did one search.

Actually I dropped the most prolific searcher, user #71845, who made over a quarter million searchs; and totally messes up my nice straight line.

Today is national mental health day. I think the most disturbing thing I’ve noticed as I browse this data is the number of people searching for information on how to commit sucide. There are effective treatments for depression.