As far as image filtering is concerned, these are commonly implemented as 'MTurk' tasks. Take a look at "Amazon Mechanical Turk" if you want this feature. You generally want to have a photo viewed 3x and get at least 2 agreements that it is a prohibited type of image. Those that disagree get a lower proficiency score and then you can exclude people with low proficiency. Most of them make a proficiency test and its a $0.00 item, but it allows the people to qualify for the paid work.
On Wednesday, November 25, 2009 10:01:34 AM UTC-7, mdipierro wrote: > > As a bad words validator you can try this one: > > http://web2py.com/examples/static/badwords.py > > Detecting p**n images is complex well beyond me. I do not know if > there is a web service that provides that as a service. > > Massimo --
