Open AccessA Secure Based Preserving Social Media Data Mangement System
Author(s) -
V. Geetha,
C. Gomathy,
Maddu Pavan Manikanta Kiran Mr.,
G. Rajesh
Publication year - 2021
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.d2455.0410421
Subject(s) - computer science , ranking (information retrieval) , rank (graph theory) , inference , social media , internet privacy , world wide web , private information retrieval , publishing , data publishing , data collection , process (computing) , information privacy , information retrieval , computer security , artificial intelligence , mathematics , combinatorics , political science , law , operating system , statistics
Personalized suggestions are important tohelp users find relevant information. It often depends on hugecollection of user data, especially users’ online activity (e.g.,liking/commenting/sharing) on social media, thereto userinterests. Publishing such user activity makes inferenceattacks easy on the users, as private data (e.g., contactdetails) are often easily gathered from the users’ activitydata. during this module, we proposed PrivacyRank, anadjustable and always protecting privacy on social media datapublishing framework , which protects users against frequentattacks while giving personal ranking basedrecommendations. Its main idea is to continuously blur useractivity data like user-specified private data is minimizedunder a given data budget, which matches round the rankingloss suffer from the knowledge blurring processso on preserve the usage of the info for enablingsuggestions. a true world evaluation on both synthetic andreal-world datasets displays that our model can provideeffective and continuous protection against to the info givenby the user, while still conserving the usage of the blurreddata for private ranking based suggestion. Compared to otherapproaches, Privacy Rank achieves both better privacyprotection and a far better usage altogether the rank basedsuggestions use cases we tested.