web logs annalysis

Options
Ornicar
Ornicar Registered Posts: 8

Hello,

I write to you again.
There are an other part I don't understand what I have to do here

Feature engineering the referrer URLs

-Split URL in referer, extracting only the hostname

-Use the Find and replace processor on referer_host, replacing t.co with twitter.com and matching on the complete value of the string

-In the same column, replace www. with an empty expression (i.e. no value), matching on substring

-Once more for referer_host, replace \..* with an empty expression, matching on regular expression. This step allows us to later put all traffic from the local Google domains under a single group.

-Reduce clutter by removing eight more columns: server_ts, referer, type, visitor_params, session_params, event_params, br_lang, and tz_off

I don't know how I can do that. Is there anyone who has a course which

Answers

Setup Info
    Tags
      Help me…