Sob! (+ new ratings algorithm)

DearCupid, posted over a year ago

I've developed a much smarter algorithm to go through the raw "vote" data and make sense of it... should set it loose next week sometime -- but the results are much, much better.

Did anyone like my new simplified header? Didn't get any comments (sob).

Have a great weekend everyone

Posted on 3 November 2006 @ 19:15 (London time) - permalink
(Log in to flag spam/offensive/junk messages to moderators)

REPLY TO MESSAGE or Start a new topic

, posted over a year ago

Andrew,

I will be the first to say...the new headers are great! It really 'cleans' up the look. When you talk about the smarter algorithm for the ratings, does that mean you've found a way to get rid of the bogus ratings, people give themselves? It really will be interesting to see what you've come up with. Will look for it next week. Have a good weekend, dear.

Irish

Posted on 3 November 2006 @ 20:59 (London time) - permalink
(Log in to flag spam/offensive/junk messages to moderators)

REPLY TO MESSAGE or Start a new topic

Sexybum, posted over a year ago

Hey guys

I like the new headings like Irish says a lot cleaner........be interesting to see the new ratings as well

Posted on 4 November 2006 @ 11:12 (London time) - permalink
(Log in to flag spam/offensive/junk messages to moderators)

REPLY TO MESSAGE or Start a new topic

Sexybum, posted over a year ago

Hey Andrew how exactly will this new filter work? Will it stop people from spamming you with poor ratings? Also will it reset all the ratings and start afresh or will any 'hoax' ratings that we may have already received still saty on our records?

Posted on 4 November 2006 @ 12:30 (London time) - permalink
(Log in to flag spam/offensive/junk messages to moderators)

REPLY TO MESSAGE or Start a new topic

DearCupid, posted over a year ago

Hi everyone

The current system has a filter that strips out the majority of bad ratings. The new filter is much smarter but more work intensive, it looks at the ratings for each user individually and works out what is good and what is bad. The current filter works on a mass group basis... The new one is much smarter and more accurate.

PS. good to see you back online SB!

Posted on 4 November 2006 @ 17:25 (London time) - permalink
(Log in to flag spam/offensive/junk messages to moderators)

REPLY TO MESSAGE or Start a new topic

Add a new message to this topic