Comments on: AOL Data - The Analysis http://wwwscope.com/2006/08/27/aol-data-the-analysis/ Keeping an eye on the World Wide Web Tue, 02 Dec 2008 12:16:37 +0000 http://wordpress.org/?v=2.6.5 By: shouting loudly » Netflix: $1m to improve our collaborative filtering http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-20 shouting loudly » Netflix: $1m to improve our collaborative filtering Mon, 02 Oct 2006 06:00:27 +0000 http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-20 [...] There’s another twist to the story: to make the contest work, they have to release the database of rental histories. Unlike the AOL search data debacle, however, Netflix carefully considered the privacy implications and got the nod from privacy experts. The data are also just easier to anonymize; a person’s web portal search records are generally much more personal than a list of rented movies. [...] [...] There’s another twist to the story: to make the contest work, they have to release the database of rental histories. Unlike the AOL search data debacle, however, Netflix carefully considered the privacy implications and got the nod from privacy experts. The data are also just easier to anonymize; a person’s web portal search records are generally much more personal than a list of rented movies. [...]

]]> By: Sean http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-15 Sean Wed, 20 Sep 2006 03:41:21 +0000 http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-15 "A surprising (to me) characteristic of this dataset was the large number of searches that were for domain names." Remember that these searches come from the AOL Client. The interface of the AOL client (in the latest version) uses a single text field for both search and URL navigation (and some AOL keywords). This may skew the results compared to users not using the AOL client. I would suspect that the number of URL searches by users not using the AOL client would be much lower. Searches for "ebay", "yahoo", and the like would probably not change much if at all. “A surprising (to me) characteristic of this dataset was the large number of searches that were for domain names.”

Remember that these searches come from the AOL Client. The interface of the AOL client (in the latest version) uses a single text field for both search and URL navigation (and some AOL keywords). This may skew the results compared to users not using the AOL client. I would suspect that the number of URL searches by users not using the AOL client would be much lower. Searches for “ebay”, “yahoo”, and the like would probably not change much if at all.

]]> By: Charles http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-10 Charles Mon, 04 Sep 2006 14:37:25 +0000 http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-10 Perhaps the "next page" events are the ones where the timestamp is identical yet shows an itemrank of 0. I noticed lots of those and thought they must be duplicates, but perhaps not. Not sure how to indicate those - I'm reluctant to go through tweaking the data having finally got it organised. (Yes, I wrote the article.) Charles Perhaps the “next page” events are the ones where the timestamp is identical yet shows an itemrank of 0. I noticed lots of those and thought they must be duplicates, but perhaps not.

Not sure how to indicate those - I’m reluctant to go through tweaking the data having finally got it organised.

(Yes, I wrote the article.)

Charles

]]> By: Nick http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-8 Nick Fri, 01 Sep 2006 09:36:24 +0000 http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-8 Ok - for those interested, the analysis is here: http://www.seo-portal.com/aol-data-analysis-i-clicks-on-search-engine-results/2006/08/09/ He's saying 11% of click-throughs are on the second search result link (not to the second page). This isn't inconsistant with my analysis - we are looking at very different things. Ok - for those interested, the analysis is here: http://www.seo-portal.com/aol-data-analysis-i-clicks-on-search-engine-results/2006/08/09/

He’s saying 11% of click-throughs are on the second search result link (not to the second page). This isn’t inconsistant with my analysis - we are looking at very different things.

]]> By: pooya http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-7 pooya Fri, 01 Sep 2006 02:18:19 +0000 http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-7 Oops, forgot to include the link to guardian's article: http://technology.guardian.co.uk/weekly/story/0,,1861112,00.html Oops, forgot to include the link to guardian’s article: http://technology.guardian.co.uk/weekly/story/0,,1861112,00.html

]]> By: Nick http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-6 Nick Thu, 31 Aug 2006 09:55:04 +0000 http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-6 Hi Pooya, I haven't seen the Guardian's analysis, and I can't find it on their site. Do you have a link for it? Even using AOL's number of "next page" queries (7,887,022/36,389,567) it still gives 21% of paging queries. The only possiblity I can see is they counted the number of individual users who paged, rather than the number of paging queries (ie they are saying that 11% of the 657,426 users use the paging feature). I haven't done any analysis to confirm than, though. Hi Pooya,

I haven’t seen the Guardian’s analysis, and I can’t find it on their site. Do you have a link for it? Even using AOL’s number of “next page” queries (7,887,022/36,389,567) it still gives 21% of paging queries.

The only possiblity I can see is they counted the number of individual users who paged, rather than the number of paging queries (ie they are saying that 11% of the 657,426 users use the paging feature). I haven’t done any analysis to confirm than, though.

]]> By: pooya http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-5 pooya Thu, 31 Aug 2006 07:55:53 +0000 http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-5 hmm, strangely Gaurdian's writer belive differently about paging queries. It seems like they used the same AOL data. What do you think? He says just 11% of people page, while you said 42%. hmm, strangely Gaurdian’s writer belive differently about paging queries. It seems like they used the same AOL data. What do you think? He says just 11% of people page, while you said 42%.

]]> By: Purposeful Procrastination http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-3 Purposeful Procrastination Wed, 30 Aug 2006 08:12:09 +0000 http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-3 <strong>Like a bad penny...</strong> Nick Lothian has shifted/diverged to another blog: WWWScope, kicking things off with a look at the AOL search data. His analysis is generally spot on, so I'm looking forward to him becoming a little more prolific (?) in the coming months. Subscribed.... Like a bad penny…

Nick Lothian has shifted/diverged to another blog: WWWScope, kicking things off with a look at the AOL search data. His analysis is generally spot on, so I’m looking forward to him becoming a little more prolific (?) in the coming months. Subscribed….

]]> By: BadMagicNumber http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-2 BadMagicNumber Mon, 28 Aug 2006 23:27:21 +0000 http://wwwscope.com/2006/08/27/aol-data-the-analysis/#comment-2 <strong>My New Blog...</strong> For a variety of reasons I've started a new blog: http://wwwscope.com/. I probably wont write much Java stuff on there (not that I seem to here, either at the moment), but my intention is that it will contain mostly longer technical content. I've st... My New Blog…

For a variety of reasons I’ve started a new blog: http://wwwscope.com/. I probably wont write much Java stuff on there (not that I seem to here, either at the moment), but my intention is that it will contain mostly longer technical content.

I’ve st…

]]>