Reducing photo-tagging spam
View Patent ↗A photo spam detector detects illegitimate non-natively captured images through extracting image features and feeding the extracted features into a probabilistic model. The probabilistic model categorizes the photo as legitimate or illegitimate. Requests to tag one or more users in a photo are analyzed by a tag analyzer that assesses relationships between the tag requests themselves, social relationships between the tagged users, and the presence or absence of faces within the regions specified by the tag requests. Based on the classification of images or tags as illegitimate, a social networking system applies one or more social media distribution policies to the image or tags to suppress or prohibit distribution.
1. A method comprising, by one or more computing systems:
receiving one or more requests to tag one or more respective users in an image, each request defining a region in the image and associating a user of the one or more users with the region;
determining a social relationship between the one or more users;
determining a probability that each of the one or more requests constitutes an illegitimate tag based on the social relationship; and
applying one or more social media distribution policies to one or more of the requests based on the probability.
2. The method of claim 1 , further comprising, for each of the one or more requests:
applying one or more facial recognition algorithms on the region defined in the request; and
increasing the determined probability if the one or more facial recognition algorithms fails to detect a face in the defined region.
3. The method of claim 2 , further comprising, for each of the one or more requests:
increasing the determined probability if the request defines a region including a face that is associated with the same user associated in a region defined by another request.
4. The method of claim 1 , further comprising:
analyzing a relationship between the one or more requests; and
adjusting the determined probability based on the analysis.
5. The method of claim 1 , further comprising adjusting the probability by decreasing the probability if requests were received in alphabetical order of user names.
6. The method of claim 1 , further comprising adjusting the probability by decreasing the probability if the requests were received at a rate above a predetermined threshold.
7. The method of claim 1 , further comprising adjusting the probability by decreasing the probability if a coefficient score between the one or more users is below a predetermined threshold.
8. The method of claim 1 , wherein one of the social media distribution policies comprises suppressing distribution of the image.
9. The method of claim 1 , wherein one of the social media distribution policies comprises suppressing distribution of actions on the image.