A system identifies an image and determines whether the image contains inappropriate content based on first data associated with the image, second data associated with a document that contains the image or refers to the image, and/or third data associated with a group of documents with which the image is associated.
1. A method performed by one or more server devices, the method comprising:
representing, by the one or more server devices, a web site using one or more sets of statistics;
generating, by the one or more server devices and using the one or more sets of statistics, a score for an image that is associated with the web site, where the generating is based on:
a color associated with the image, where the color is associated with a skin tone, and
a distribution of the color across one or more other images associated with the web site;
associating, by the one or more server devices, the score with the image; and
using, by the one or more server devices, the associated score to determine whether to provide the image to a user.
2. The method of claim 1 , where the image is associated with a first document of the web site, and where the one or more other images are associated with one or more second documents of the web site, where the first document is different from the one or more second documents.
3. The method of claim 1 , where generating the score is further based on text in a document that includes the image.
4. The method of claim 3 , where generating the score includes:
generating a text score for the document based on the text in the document, the text score representing a likelihood that the document includes inappropriate content, and
using the text score when generating the score for the image.
5. The method of claim 1 , further comprising:
determining not to provide the image to the user when the associated score exceeds a threshold.
6. A system, comprising:
one or more server devices to:
represent a web site using one or more sets of statistics;
generate, using the one or more sets of statistics, a score for an image, that is associated with the web site, based on:
a color associated with the image, where the color is associated with a skin tone, and
a distribution of the color across one or more other images associated with the web site;
associate the score with the image; and
use the associated score to determine whether to provide the image to a user.
7. The system of claim 6 , where the image is associated with a first document of the web site, and where the one or more other images are associated with one or more second documents of the web site, where the first document is different from the one or more second documents.
8. The system of claim 6 , where the score is further based on text in a document that includes the image.
9. The system of claim 8 , where when generating the score, the one or more server devices are to:
generate a text score for the document based on the text in the document, the text score representing a likelihood that the document includes inappropriate content, and
use the text score when generating the score for the image.
10. The system of claim 6 , where the one or more server devices are further to:
determine not to provide the image to the user when the associated score exceeds a threshold.
11. A non-transitory computer-readable memory device having computer-executable instructions stored thereon, the computer-executable instructions comprising:
one or more instructions to represent a web site using one or more sets of statistics;
one or more instructions to generate, using the one or more sets of statistics, a score for an image, that is associated with the web site, based on:
a color associated with the image, where the color is associated with a skin tone, and
a distribution of the color across one or more other images associated with the web site;
one or more instructions to associate the score with the image; and
one or more instructions to use the associated score to determine whether to provide the image to a user.
12. The non-transitory computer-readable memory device of claim 11 , where the image is associated with a first document of the web site, and where the one or more other images are associated with one or more second documents of the web site, where the first document is different from the one or more second documents.
13. The non-transitory computer-readable memory device of claim 11 , where the score is further based on text in a document that includes the image.
14. The non-transitory computer-readable memory device of claim 13 , where the one or more instructions to generate the score include:
one or more instructions to generate a text score for the document based on the text in the document, the text score representing a likelihood that the document includes inappropriate content, and
one or more instructions to use the text score when generating the score for the image.
15. The non-transitory computer-readable memory device of claim 11 , the computer-executable instructions further comprising:
one or more instructions to determine not to provide the image to the user when the associated score exceeds a threshold.