Microblogging and Advertisting Data Mining Challenge

Completed • $8,000 • 163 teams

KDD Cup 2012, Track 2

Mon 20 Feb 2012
– Fri 1 Jun 2012 (2 years ago)

Same tokensid in different *_tokensid.txt files

« Prev
Topic
» Next
Topic

Do they have the same meaning?

For example, you can find tokensid "75" in the 4th line of purchasedkeywordid_tokensid.txt, 2nd line of queryid_tokensid.txt, and 1st line of titleid_tokensid.txt. Do these "75"s mean the same word?

good question. this is important

yes,i think they are the same word,because in "purchasedkeywordidtokensid.txt" every line map one ad purchased keywords, and if a user issue a query contains the keyword,he will see the ad! so the same tokenid in "queryidtokensid.txt" or "titleid_tokensid.txt" may means the same word

Does intersection of ad purchased keywords and query sets never empty?

Hi,

I differ from the opinion discussed in the thread! This is because ids of keyword or titile or any other doesn't have any meaning(atleast in my understaning).It is the list of tokens that it points out has some underlying meaning.If we could find the intersection of the tokens then i believe it would help us in some way i believe.Please comment on this

Thanks

There is no easy way to determine if the numbers representing tokens are specific to each variable (query, keyword, title, description) or common across all four variables.  Maybe the competition sponsors could enlighten us on this issue.

Reply

Flag alert Flagging is a way of notifying administrators that this message contents inappropriate or abusive content. Are you sure this forum post qualifies?