An A3P approach towards Image Privacy policy recommendation on content sharing sites

Popularity of the social media like Facebook, Flickr, LinkedIn and others has increasing everyday due its various features provided. Some of the features are to collaborate with friends, good and friendly user interface to share data and multimedia content at ease. Data uploaded to Social network by the user are at high risk of vulnerability. User uploaded data on the social media does play vital role in user connectivity. Thus maintaining privacy of the user uploaded data on content distribution network is at most important. To control privacy of the uploaded data user should also provide privacy information for the uploaded data. It is observed the difficulty in setting privacy for images shared by user in social network. This requires automation privacy recommendation for the user shared images based on the image metadata. This suggested privacy inference for uploaded image should suits his satisfaction level. The necessary aspect considered are social context and image metadata for analyzing the privacy setting for the user uploaded data. Two-level framework is proposed which utilizes user’s history for recommending best policy setting for uploaded image. Solution provided in paper relies on an categorizing the image based on image metadata which may belong to similar policies and predicting the policy using apriori data mining algorithm to automatically suggest a policy for each image uploaded by the user in social network. The goal of this paper is to provide extensive method for privacy policy suggestions to improve the security of shared data in the social media sites. Keywords— Content sharing sites, Apriori data mining algorithm, Social media, privacy inference.